diff --git a/CHANGELOG.md b/CHANGELOG.md
index e273c11..e17132d 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,14 @@
+## [3.0.2](https://github.com/jmlweb/hyntx/compare/v3.0.1...v3.0.2) (2026-03-22)
+
+### Bug Fixes
+
+- **security:** remove ajv override that broke ESLint (ajv v6/v8 incompatibility) ([88aac5c](https://github.com/jmlweb/hyntx/commit/88aac5cce1820ec62c1c1ec0f90c45e3b4004767))
+- update dependency overrides to resolve security vulnerabilities ([685befa](https://github.com/jmlweb/hyntx/commit/685befaffa058c5e6b45fd57432c73b0664aad29))
+
+### Documentation
+
+- add quality assessment report ([722de83](https://github.com/jmlweb/hyntx/commit/722de83b18ee4f332d79c02fe69b7cf03960038d))
+
 ## [3.0.1](https://github.com/jmlweb/hyntx/compare/v3.0.0...v3.0.1) (2026-01-27)
 
 ### Bug Fixes
diff --git a/README.md b/README.md
index d67e0b9..c50a740 100644
--- a/README.md
+++ b/README.md
@@ -176,7 +176,7 @@ hyntx -m individual               # Short form
 - Analyzing high-stakes or complex prompts
 - Conducting quality audits or teaching sessions
 
-**Performance Note**: Numbers based on `gemma3:4b` on CPU. Actual speed varies by hardware, model size, and prompt complexity.
+**Performance Note**: Numbers based on `gemma4:e4b` on CPU. Actual speed varies by hardware, model size, and prompt complexity.
 
 **Detailed Guide**: See [Analysis Modes Documentation](./docs/ANALYSIS_MODES.md) for comprehensive comparison, examples, and decision guidelines.
 
@@ -255,17 +255,17 @@ Configure one or more providers in priority order. Hyntx will try each provider
 ```bash
 # Single provider (Ollama only)
 export HYNTX_SERVICES=ollama
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 
 # Multi-provider with fallback (tries Ollama first, then Anthropic)
 export HYNTX_SERVICES=ollama,anthropic
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 export HYNTX_ANTHROPIC_KEY=sk-ant-your-key-here
 
 # Cloud-first with local fallback
 export HYNTX_SERVICES=anthropic,ollama
 export HYNTX_ANTHROPIC_KEY=sk-ant-your-key-here
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 ```
 
 #### Provider-Specific Variables
@@ -274,7 +274,7 @@ export HYNTX_OLLAMA_MODEL=gemma3:4b
 
 | Variable             | Default                  | Description       |
 | -------------------- | ------------------------ | ----------------- |
-| `HYNTX_OLLAMA_MODEL` | `gemma3:4b`              | Model to use      |
+| `HYNTX_OLLAMA_MODEL` | `gemma4:e4b`             | Model to use      |
 | `HYNTX_OLLAMA_HOST`  | `http://localhost:11434` | Ollama server URL |
 
 **Anthropic:**
@@ -303,7 +303,7 @@ export HYNTX_REMINDER=7d
 ```bash
 # Add to ~/.zshrc or ~/.bashrc (or let Hyntx auto-save it)
 export HYNTX_SERVICES=ollama,anthropic
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 export HYNTX_ANTHROPIC_KEY=sk-ant-your-key-here
 export HYNTX_REMINDER=14d
 
@@ -327,7 +327,7 @@ Ollama runs AI models locally for **privacy and cost savings**.
 2. Pull a model:
 
    ```bash
-   ollama pull gemma3:4b
+   ollama pull gemma4:e4b
    ```
 
 3. Verify it's running:
@@ -369,7 +369,7 @@ Configure multiple providers for automatic fallback:
 ```bash
 # If Ollama is down, automatically try Anthropic
 export HYNTX_SERVICES=ollama,anthropic
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 export HYNTX_ANTHROPIC_KEY=sk-ant-your-key-here
 ```
 
@@ -481,11 +481,11 @@ If using Ollama (recommended for privacy):
 ollama serve
 
 # Pull a model if needed
-ollama pull gemma3:4b
+ollama pull gemma4:e4b
 
 # Set environment variables (add to ~/.zshrc or ~/.bashrc)
 export HYNTX_SERVICES=ollama
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 ```
 
 ### Available MCP Tools
@@ -662,7 +662,7 @@ Use check-context to verify: "Update the component to handle errors"
 #### "Slow responses"
 
 - Local Ollama models are fastest but require GPU for best performance
-- Consider using a faster model: `export HYNTX_OLLAMA_MODEL=gemma3:4b:1b`
+- Consider using a faster model: `export HYNTX_OLLAMA_MODEL=gemma4:e2b`
 - Cloud providers (Anthropic, Google) offer faster responses but require API keys
 
 ## Privacy & Security
@@ -708,15 +708,15 @@ For local analysis with Ollama, you need to have a compatible model installed. S
 
 | Use Case            | Model         | Parameters | Disk Size | Speed (CPU)    | Quality   |
 | ------------------- | ------------- | ---------- | --------- | -------------- | --------- |
-| **Daily use**       | `gemma3:4b`   | 2-3B       | ~2GB      | ~2-5s/prompt   | Good      |
+| **Daily use**       | `gemma4:e4b`  | ~5GB Q4    | ~5GB      | ~3-7s/prompt   | Good      |
 | **Production**      | `mistral:7b`  | 7B         | ~4GB      | ~5-10s/prompt  | Better    |
 | **Maximum quality** | `qwen2.5:14b` | 14B        | ~9GB      | ~15-30s/prompt | Excellent |
 
 **Installation**:
 
 ```bash
-# Install recommended model (gemma3:4b)
-ollama pull gemma3:4b
+# Install recommended model (gemma4:e4b)
+ollama pull gemma4:e4b
 
 # Or choose a different model
 ollama pull mistral:7b
diff --git a/docs/DATA_PROCESSING_ANALYSIS.md b/docs/DATA_PROCESSING_ANALYSIS.md
new file mode 100644
index 0000000..e69de29
diff --git a/docs/HEURISTICS_ANALYSIS.md b/docs/HEURISTICS_ANALYSIS.md
new file mode 100644
index 0000000..e43dfe0
--- /dev/null
+++ b/docs/HEURISTICS_ANALYSIS.md
@@ -0,0 +1,91 @@
+# Heuristics Analysis Report
+
+**Date:** 2026-01-30
+**Author:** HAL (assisted analysis)
+
+## Overview
+
+This document analyzes the `extractRealExamples()` function in `src/core/aggregator.ts` and the category mapping in `src/providers/base.ts`.
+
+## Current Architecture
+
+### Analysis Pipeline
+
+```
+Prompts → AI Provider → Minimal/Individual Result → Aggregator → Full AnalysisResult → Semantic Validator
+```
+
+### Key Components
+
+1. **ISSUE_TAXONOMY** (`schemas.ts`): 8 predefined issue types
+   - `vague`, `no-context`, `too-broad`, `no-goal`, `imperative`
+   - `missing-technical-details`, `unclear-priorities`, `insufficient-constraints`
+
+2. **extractRealExamples()** (`aggregator.ts`): Heuristic matcher for fallback examples
+   - Only used when AI doesn't provide examples (minimal mode)
+   - Uses boolean matching with specific patterns per issue type
+
+3. **Individual Mode**: AI returns per-prompt results with real examples
+   - `parseBatchIndividualResponse()` in `base.ts`
+   - Examples come directly from AI categorization
+
+## Findings
+
+### extractRealExamples() Heuristics
+
+The current implementation uses strict boolean matching:
+
+| Issue Type | Current Heuristic                                                         |
+| ---------- | ------------------------------------------------------------------------- |
+| vague      | < 50 chars, ≤ 5 words, generic verbs, no file extensions                  |
+| no-context | Has pronouns (this/it/that), no files, no function/component/method/class |
+| too-broad  | > 100 chars, ≥ 2 "and", has also/then/build/create                        |
+| no-goal    | < 30 chars, ≤ 4 words, no action verbs, no question mark                  |
+| imperative | < 20 chars, ≤ 3 words, starts with verb                                   |
+
+### Category Mapping Inconsistency
+
+`base.ts` uses different category IDs than `schemas.ts`:
+
+| base.ts (individual mode) | schemas.ts (taxonomy) |
+| ------------------------- | --------------------- |
+| `vague-request`           | `vague`               |
+| `missing-context`         | `no-context`          |
+| `unclear-goal`            | `no-goal`             |
+
+## Recommendations
+
+### 1. Unify Category IDs
+
+Add mapping in `base.ts`:
+
+```typescript
+const CATEGORY_TO_TAXONOMY_ID: Record<string, string> = {
+  'vague-request': 'vague',
+  'missing-context': 'no-context',
+  'unclear-goal': 'no-goal',
+  // ... etc
+};
+```
+
+### 2. Improve Heuristics (Future Work)
+
+Consider scoring-based matching instead of boolean:
+
+- Calculate match score (0-1) per prompt per issue
+- Select highest-scoring examples
+- More nuanced matching for edge cases
+
+### 3. Individual Mode Already Works Well
+
+The individual/batch-individual schema already extracts real examples from AI responses. The heuristics in `extractRealExamples()` are only a fallback for minimal mode.
+
+## Test Coverage
+
+- `aggregator.test.ts`: 50 tests, all passing
+- Tests cover all issue types and edge cases
+- Gold standard in `benchmark/gold-standard.ts`: 50 prompts across 4 tiers
+
+## Conclusion
+
+The current architecture is solid. The main improvement opportunity is unifying category mappings between individual mode and the taxonomy. The heuristics work correctly for their intended purpose as a fallback.
diff --git a/docs/MINIMUM_VIABLE_MODEL.md b/docs/MINIMUM_VIABLE_MODEL.md
index 0f0879e..18f447c 100644
--- a/docs/MINIMUM_VIABLE_MODEL.md
+++ b/docs/MINIMUM_VIABLE_MODEL.md
@@ -2,13 +2,13 @@
 
 ## Executive Summary
 
-**Minimum viable model: `gemma3:4b` (2-3B parameters, ~2GB disk, ~2-5s/prompt CPU)**
+**Minimum viable model: `gemma4:e4b` (~5GB Q4, 128K context, native function calling, ~3-7s/prompt CPU)**
 
 This document documents the findings from the analysis to determine the minimum viable Ollama model that can generate valid and useful results with Hyntx.
 
 **Quick recommendations**:
 
-- **Minimal viable**: `gemma3:4b` (2B) - Fast, lightweight, good for daily use
+- **Minimal viable**: `gemma4:e4b` (~5GB Q4) - 128K context, native function calling, good for daily use
 - **Production quality**: `mistral:7b` (7B) - Better analysis, moderate resources
 - **Maximum quality**: `qwen2.5:14b` or `llama3:70b` - Full schema, requires GPU for 70B
 
@@ -33,16 +33,16 @@ Hyntx uses an adaptive system that adjusts the analysis schema based on model si
 
 ### Models Tested
 
-| Model       | Parameters | Disk Size | Schema  | Result   | Quality   | Speed (CPU)  |
-| ----------- | ---------- | --------- | ------- | -------- | --------- | ------------ |
-| `gemma3:4b` | 2-3B       | ~2GB      | Minimal | ✅ Works | Excellent | ~2-5s/prompt |
-| `gemma3:4b` | 4B         | ~3.3GB    | Minimal | ✅ Works | Excellent | ~3-6s/prompt |
+| Model        | Parameters | Disk Size | Schema  | Result   | Quality   | Speed (CPU)  |
+| ------------ | ---------- | --------- | ------- | -------- | --------- | ------------ |
+| `gemma4:e4b` | 2-3B       | ~2GB      | Minimal | ✅ Works | Excellent | ~2-5s/prompt |
+| `gemma4:e4b` | 4B         | ~3.3GB    | Minimal | ✅ Works | Excellent | ~3-6s/prompt |
 
 ### Quality Analysis
 
 **Test performed**: Analysis of 52 prompts from current day
 
-**Results with `gemma3:4b`**:
+**Results with `gemma4:e4b`**:
 
 - ✅ Valid JSON generated correctly
 - ✅ Valid and consistent issue IDs (no-context, vague, too-broad, imperative)
@@ -51,9 +51,9 @@ Hyntx uses an adaptive system that adjusts the analysis schema based on model si
 - ✅ Reasonable scores (0-100 scale)
 - ✅ No parsing errors
 
-**Results with `gemma3:4b`**:
+**Results with `gemma4:e4b`**:
 
-- ✅ Identical results to `gemma3:4b`
+- ✅ Identical results to `gemma4:e4b`
 - ✅ Same quality and consistency
 - ✅ No notable differences
 
@@ -71,21 +71,22 @@ Models with fewer parameters or poor instruction-following capabilities will hav
 
 ## Recommended Minimum Model
 
-### `gemma3:4b` (default)
+### `gemma4:e4b` (default)
 
 **Reasons**:
 
 - ✅ It's the system default model
-- ✅ Classified as "micro" (automatically uses minimal schema)
+- ✅ Uses `small` strategy which maps to **Full Schema** — complete analysis with patterns, examples, and before/after
 - ✅ Works perfectly in real tests
-- ✅ Reasonable balance between size and capability
-- ✅ Manageable size (~2GB on disk)
+- ✅ 128K context window handles large prompt batches without truncation
+- ✅ Native function calling enables more reliable structured output
+- ✅ ~5GB Q4 — good balance between size and capability
 - ✅ Acceptable speed on CPU/GPU
 
 **Configuration**:
 
 ```bash
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 export HYNTX_OLLAMA_HOST=http://localhost:11434
 ```
 
@@ -97,7 +98,7 @@ export HYNTX_OLLAMA_HOST=http://localhost:11434
 
 **Why it's the balanced choice**:
 
-- ✅ **Better quality**: Uses the "Small Schema" (more detailed than Minimal Schema used by `gemma3:4b`)
+- ✅ **Better quality**: Uses the "Small Schema" (comparable to the Full Schema used by `gemma4:e4b`)
   - Better analysis quality with pattern detection and basic analysis
   - Some custom examples extracted from your prompts
   - Basic contextual information included
@@ -119,7 +120,7 @@ export HYNTX_OLLAMA_MODEL=mistral:7b
 
 **When to use `mistral:7b`**:
 
-- You want better analysis quality than `gemma3:4b` but don't need maximum quality
+- You want a 7B model with comparable schema coverage to `gemma4:e4b` but prefer mistral's output style
 - You have modern hardware (8GB+ RAM, modern CPU)
 - You're doing production analysis or code reviews
 - You want custom examples from your prompts (not just taxonomy-based examples)
@@ -134,12 +135,12 @@ export HYNTX_OLLAMA_MODEL=mistral:7b
 
 ### Micro Models (Minimal Schema)
 
-| Model       | Parameters | Disk Size | Speed (CPU)  | Status                   |
-| ----------- | ---------- | --------- | ------------ | ------------------------ |
-| `gemma3:4b` | 2-3B       | ~2GB      | ~2-5s/prompt | ✅ Recommended (default) |
-| `gemma3:4b` | 4B         | ~3.3GB    | ~3-6s/prompt | ✅ Tested, works well    |
-| `phi3:mini` | 3.8B       | ~2.3GB    | ~3-5s/prompt | Expected to work         |
-| `gemma2:2b` | 2B         | ~1.6GB    | ~1-3s/prompt | Theoretically viable     |
+| Model        | Parameters | Disk Size | Speed (CPU)  | Status                   |
+| ------------ | ---------- | --------- | ------------ | ------------------------ |
+| `gemma4:e4b` | 2-3B       | ~2GB      | ~2-5s/prompt | ✅ Recommended (default) |
+| `gemma4:e4b` | 4B         | ~3.3GB    | ~3-6s/prompt | ✅ Tested, works well    |
+| `phi3:mini`  | 3.8B       | ~2.3GB    | ~3-5s/prompt | Expected to work         |
+| `gemma2:2b`  | 2B         | ~1.6GB    | ~1-3s/prompt | Theoretically viable     |
 
 ### Small Models (Small Schema - Better Quality)
 
@@ -201,17 +202,17 @@ For better quality, use models that support full schema (≥ 8B parameters).
 
 ## Usage Recommendations
 
-### For Development/Testing (Minimal Schema)
+### For Development/Testing (Full Schema)
 
 ```bash
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 ```
 
-- **Parameters**: 2-3B
-- **Speed**: ~2-5s/prompt (CPU)
+- **Size**: ~5GB Q4
+- **Speed**: ~3-7s/prompt (CPU)
 - **Use case**: Fast iteration, daily use
-- ✅ Valid and useful results
-- ✅ Lightweight and fast
+- ✅ Full schema — complete patterns, examples, before/after
+- ✅ 128K context, native function calling
 
 ### For Professional Analysis (Small Schema)
 
@@ -252,7 +253,7 @@ ollama list
 
 # Test with Hyntx
 export HYNTX_SERVICES=ollama
-export HYNTX_OLLAMA_MODEL=gemma3:4b
+export HYNTX_OLLAMA_MODEL=gemma4:e4b
 hyntx --date today --output test.json
 
 # Verify valid JSON
@@ -278,24 +279,24 @@ If the command generates valid JSON with patterns, the model is viable.
 
 ## Conclusion
 
-**The confirmed minimum viable model is `gemma3:4b` (2-3B parameters, ~2GB disk)**.
+**The confirmed minimum viable model is `gemma4:e4b` (~5GB Q4, 128K context, native function calling)**.
 
 This model:
 
-- ✅ Works correctly with the minimal schema
+- ✅ Works correctly with the full schema (`small` strategy → full schema)
 - ✅ Generates valid and useful results
 - ✅ Is the system default
 - ✅ Provides optimal balance between size, speed, and quality
-- ✅ Fast enough for daily use (~2-5s/prompt on CPU)
+- ✅ Fast enough for daily use (~3-7s/prompt on CPU)
 
 **Recommendations by use case**:
 
-- **Daily development**: `gemma3:4b` (2-3B) - Minimal schema
+- **Daily development**: `gemma4:e4b` (~5GB Q4) - Full schema (128K context, native function calling)
 - **Production analysis**: `mistral:7b` (7B) - Small schema
 - **Team retrospectives**: `qwen2.5:14b` (14B) - Full schema
 - **Maximum quality**: `llama3:70b` (70B) - Full schema (GPU needed)
 
-Most users will find `gemma3:4b` sufficient. For deeper analysis, use models ≥ 7B parameters that support small or full schemas.
+Most users will find `gemma4:e4b` sufficient — it uses the full schema with 128K context and native function calling. For even deeper analysis, use models ≥ 14B parameters.
 
 ## Benchmark Results (2026-01-27)
 
@@ -309,22 +310,22 @@ Most users will find `gemma3:4b` sufficient. For deeper analysis, use models ≥
 
 | Model        | Time   | Score | Patterns | Status                    |
 | ------------ | ------ | ----- | -------- | ------------------------- |
-| gemma3:4b    | 44s ⚡ | 6/10  | 5        | ✅ **Best choice**        |
+| gemma4:e4b   | 44s ⚡ | 6/10  | 5        | ✅ **Best choice**        |
 | codellama:7b | 82s    | 8/10  | 1        | ❌ Returns placeholders   |
 | mistral:7b   | 89s    | 4/10  | 5        | ✅ Good                   |
-| gemma3:4b    | 207s   | 6/10  | 5        | ⚠️ Slow, had counting bug |
+| gemma4:e4b   | 207s   | 6/10  | 5        | ⚠️ Slow, had counting bug |
 
 ### Key Findings
 
-1. **gemma3:4b is the recommended default** - 4x faster than gemma3:4b with better results
+1. **gemma4:e4b is the recommended default** - 4x faster than gemma4:e4b with better results
 2. **codellama:7b is NOT recommended** - Returns placeholder text instead of real analysis
-3. **gemma3:4b has bugs** - Reported 44 prompts when only 9 were analyzed (fixed in code)
-4. **mistral:7b is reliable** - Good quality but slower than gemma3:4b
+3. **gemma4:e4b has bugs** - Reported 44 prompts when only 9 were analyzed (fixed in code)
+4. **mistral:7b is reliable** - Good quality but slower than gemma4:e4b
 
 ### Recommendation Update
 
 Based on these benchmarks, the recommended models are:
 
-1. **Daily use**: `gemma3:4b` (fast, accurate)
+1. **Daily use**: `gemma4:e4b` (fast, accurate)
 2. **Fallback**: `mistral:7b` (slower but reliable)
 3. **Avoid**: `codellama:7b` (returns placeholder text)
diff --git a/docs/QUALITY_ASSESSMENT.md b/docs/QUALITY_ASSESSMENT.md
index 64ce3b4..c6689f9 100644
--- a/docs/QUALITY_ASSESSMENT.md
+++ b/docs/QUALITY_ASSESSMENT.md
@@ -24,7 +24,7 @@ This report documents a comprehensive quality assessment of Hyntx's analysis out
 | Model          | Quality       | Speed | Recommendation                             |
 | -------------- | ------------- | ----- | ------------------------------------------ |
 | `mistral:7b`   | ✅ Good       | 89s   | **Recommended for production**             |
-| `gemma3:4b`    | ⚠️ Acceptable | 44s   | Good for quick analysis, some placeholders |
+| `gemma4:e4b`   | ⚠️ Acceptable | 44s   | Good for quick analysis, some placeholders |
 | `llama3.2`     | ⚠️ Acceptable | 207s  | Slow, functional                           |
 | `codellama:7b` | ❌ Unusable   | 82s   | Returns only placeholder text              |
 
@@ -53,7 +53,7 @@ This report documents a comprehensive quality assessment of Hyntx's analysis out
 }
 ```
 
-#### gemma3:4b (Default)
+#### gemma4:e4b (Default)
 
 **Strengths:**
 
@@ -150,7 +150,7 @@ Placeholder detection doesn't catch all cases:
 
 - **Date analyzed:** 2026-01-02
 - **Project:** hyntx
-- **Model:** gemma3:4b
+- **Model:** gemma4:e4b
 - **Prompts found:** 9
 - **Successfully analyzed:** 4 (44%)
 - **Skipped due to errors:** 5
@@ -202,7 +202,7 @@ However, users should be aware that:
 1. Output quality varies significantly by model
 2. Some bugs exist in stats calculation
 3. `codellama:7b` should be avoided entirely
-4. The default `gemma3:4b` is fast but may include some placeholder text
+4. The default `gemma4:e4b` is fast but may include some placeholder text
 
 The project status has been updated from "NOT READY FOR USE" to "BETA" to reflect its current functional state.
 
diff --git a/package.json b/package.json
index d50790d..a46a5d9 100644
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "hyntx",
-  "version": "3.0.1",
+  "version": "3.0.2",
   "description": "CLI that analyzes Claude Code prompts and generates improvement suggestions",
   "type": "module",
   "packageManager": "pnpm@9.15.4",
@@ -70,7 +70,7 @@
     "husky": "^9.1.7",
     "lint-staged": "^16.2.7",
     "prettier": "^3.7.4",
-    "semantic-release": "^25.0.2",
+    "semantic-release": "^25.0.3",
     "tsup": "^8.5.1",
     "typescript": "^5.9.3",
     "typescript-eslint": "^8.51.0",
@@ -129,10 +129,23 @@
   },
   "pnpm": {
     "overrides": {
-      "hono": ">=4.11.4",
-      "lodash": ">=4.17.23",
-      "lodash-es": ">=4.17.23",
-      "undici": ">=7.18.2"
+      "hono": ">=4.12.4",
+      "lodash": ">=4.18.0",
+      "lodash-es": ">=4.18.0",
+      "undici": ">=7.24.0",
+      "@hono/node-server": ">=1.19.10",
+      "@modelcontextprotocol/sdk": ">=1.26.0",
+      "@isaacs/brace-expansion": ">=5.0.5",
+      "minimatch": ">=10.2.3",
+      "rollup": ">=4.59.0",
+      "flatted": ">=3.4.2",
+      "qs": ">=6.14.2",
+      "handlebars": ">=4.7.9",
+      "picomatch": ">=4.0.4",
+      "brace-expansion": ">=5.0.5",
+      "path-to-regexp": ">=8.4.0",
+      "vite": ">=7.3.2",
+      "yaml": ">=2.8.3"
     }
   }
 }
diff --git a/pnpm-lock.yaml b/pnpm-lock.yaml
index b762384..d2c0ed7 100644
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -5,18 +5,31 @@ settings:
   excludeLinksFromLockfile: false
 
 overrides:
-  hono: '>=4.11.4'
-  lodash: '>=4.17.23'
-  lodash-es: '>=4.17.23'
-  undici: '>=7.18.2'
+  hono: '>=4.12.4'
+  lodash: '>=4.18.0'
+  lodash-es: '>=4.18.0'
+  undici: '>=7.24.0'
+  '@hono/node-server': '>=1.19.10'
+  '@modelcontextprotocol/sdk': '>=1.26.0'
+  '@isaacs/brace-expansion': '>=5.0.5'
+  minimatch: '>=10.2.3'
+  rollup: '>=4.59.0'
+  flatted: '>=3.4.2'
+  qs: '>=6.14.2'
+  handlebars: '>=4.7.9'
+  picomatch: '>=4.0.4'
+  brace-expansion: '>=5.0.5'
+  path-to-regexp: '>=8.4.0'
+  vite: '>=7.3.2'
+  yaml: '>=2.8.3'
 
 importers:
 
   .:
     dependencies:
       '@modelcontextprotocol/sdk':
-        specifier: ^1.25.3
-        version: 1.25.3(hono@4.11.7)(zod@4.3.6)
+        specifier: '>=1.26.0'
+        version: 1.27.1(zod@4.3.6)
       asciichart:
         specifier: ^1.5.25
         version: 1.5.25
@@ -80,19 +93,19 @@ importers:
         version: 1.0.5
       '@jmlweb/tsup-config-base':
         specifier: ^1.1.4
-        version: 1.1.4(tsup@8.5.1(jiti@2.6.1)(postcss@8.5.6)(typescript@5.9.3)(yaml@2.8.2))
+        version: 1.1.4(tsup@8.5.1(jiti@2.6.1)(postcss@8.5.8)(typescript@5.9.3)(yaml@2.8.3))
       '@jmlweb/vitest-config':
         specifier: ^2.0.0
-        version: 2.0.0(vitest@4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2))
+        version: 2.0.0(vitest@4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3))
       '@semantic-release/changelog':
         specifier: ^6.0.3
-        version: 6.0.3(semantic-release@25.0.2(typescript@5.9.3))
+        version: 6.0.3(semantic-release@25.0.3(typescript@5.9.3))
       '@semantic-release/exec':
         specifier: ^7.1.0
-        version: 7.1.0(semantic-release@25.0.2(typescript@5.9.3))
+        version: 7.1.0(semantic-release@25.0.3(typescript@5.9.3))
       '@semantic-release/git':
         specifier: ^10.0.1
-        version: 10.0.1(semantic-release@25.0.2(typescript@5.9.3))
+        version: 10.0.1(semantic-release@25.0.3(typescript@5.9.3))
       '@types/asciichart':
         specifier: ^1.5.8
         version: 1.5.8
@@ -107,7 +120,7 @@ importers:
         version: 2.4.9
       '@vitest/coverage-v8':
         specifier: ^4.0.16
-        version: 4.0.18(vitest@4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2))
+        version: 4.0.18(vitest@4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3))
       conventional-changelog-conventionalcommits:
         specifier: ^9.1.0
         version: 9.1.0
@@ -130,11 +143,11 @@ importers:
         specifier: ^3.7.4
         version: 3.8.1
       semantic-release:
-        specifier: ^25.0.2
-        version: 25.0.2(typescript@5.9.3)
+        specifier: ^25.0.3
+        version: 25.0.3(typescript@5.9.3)
       tsup:
         specifier: ^8.5.1
-        version: 8.5.1(jiti@2.6.1)(postcss@8.5.6)(typescript@5.9.3)(yaml@2.8.2)
+        version: 8.5.1(jiti@2.6.1)(postcss@8.5.8)(typescript@5.9.3)(yaml@2.8.3)
       typescript:
         specifier: ^5.9.3
         version: 5.9.3
@@ -143,7 +156,7 @@ importers:
         version: 8.54.0(eslint@9.39.2(jiti@2.6.1))(typescript@5.9.3)
       vitest:
         specifier: ^4.0.16
-        version: 4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2)
+        version: 4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3)
 
 packages:
 
@@ -257,6 +270,15 @@ packages:
     resolution: {integrity: sha512-VmIFV/JkBRhDRRv7N5B7zEUkNZIx9Mp+8Pe65erz0rKycXLsi8Epcw0XJ+btSeRXgTzE7DyOyA9bkJ9mn/yqVQ==}
     engines: {node: '>=v18'}
 
+  '@emnapi/core@1.9.2':
+    resolution: {integrity: sha512-UC+ZhH3XtczQYfOlu3lNEkdW/p4dsJ1r/bP7H8+rhao3TTTMO1ATq/4DdIi23XuGoFY+Cz0JmCbdVl0hz9jZcA==}
+
+  '@emnapi/runtime@1.9.2':
+    resolution: {integrity: sha512-3U4+MIWHImeyu1wnmVygh5WlgfYDtyf0k8AbLhMFxOipihf6nrWC4syIm/SwEeec0mNSafiiNnMJwbza/Is6Lw==}
+
+  '@emnapi/wasi-threads@1.2.1':
+    resolution: {integrity: sha512-uTII7OYF+/Mes/MrcIOYp5yOtSMLBWSIoLPpcgwipoiKbli6k322tcoFsxoIIxPDqW01SQGAgko4EzZi2BNv2w==}
+
   '@esbuild/aix-ppc64@0.27.2':
     resolution: {integrity: sha512-GZMB+a0mOMZs4MpDbj8RJp4cw+w1WV5NYD6xzgvzUJ5Ek2jerwfO2eADyI6ExDSUED+1X8aMbegahsJi+8mgpw==}
     engines: {node: '>=18'}
@@ -451,11 +473,11 @@ packages:
     resolution: {integrity: sha512-43/qtrDUokr7LJqoF2c3+RInu/t4zfrpYdoSDfYyhg52rwLV6TnOvdG4fXm7IkSB3wErkcmJS9iEhjVtOSEjjA==}
     engines: {node: ^18.18.0 || ^20.9.0 || >=21.1.0}
 
-  '@hono/node-server@1.19.9':
-    resolution: {integrity: sha512-vHL6w3ecZsky+8P5MD+eFfaGTyCeOHUIFYMGpQGbrBTSmNNoxv0if69rEZ5giu36weC5saFuznL411gRX7bJDw==}
+  '@hono/node-server@1.19.11':
+    resolution: {integrity: sha512-dr8/3zEaB+p0D2n/IUrlPF1HZm586qgJNXK1a9fhg/PzdtkK7Ksd5l312tJX2yBuALqDYBlG20QEbayqPyxn+g==}
     engines: {node: '>=18.14.1'}
     peerDependencies:
-      hono: '>=4.11.4'
+      hono: '>=4.12.4'
 
   '@humanfs/core@0.19.1':
     resolution: {integrity: sha512-5DyQ4+1JEUzejeK1JGICcideyfUbGixgS9jNgex5nqkW+cY7WZhxBigmieN5Qnw9ZosSNVC9KQKyb+GUaGyKUA==}
@@ -473,14 +495,6 @@ packages:
     resolution: {integrity: sha512-bV0Tgo9K4hfPCek+aMAn81RppFKv2ySDQeMoSZuvTASywNTnVJCArCZE2FWqpvIatKu7VMRLWlR1EazvVhDyhQ==}
     engines: {node: '>=18.18'}
 
-  '@isaacs/balanced-match@4.0.1':
-    resolution: {integrity: sha512-yzMTt9lEb8Gv7zRioUilSglI0c0smZ9k5D65677DLWLtWJaXIS3CqcGyUFByYKlnUj6TkjLVs54fBl6+TiGQDQ==}
-    engines: {node: 20 || >=22}
-
-  '@isaacs/brace-expansion@5.0.0':
-    resolution: {integrity: sha512-ZT55BDLV0yv0RBm2czMiZ+SqCGO7AvmOM3G/w2xhVPH+te0aKgFjmBvGlL1dH+ql2tgGO3MVrbb3jCKyvpgnxA==}
-    engines: {node: 20 || >=22}
-
   '@jmlweb/commitlint-config@3.0.1':
     resolution: {integrity: sha512-yZMn8spTCuv6n2O9JV3imMAW+ADPwS/snDEIMR+0wHC7kv83joVwFkLBuwzvPV1o9TtE3PxWqZopwxftpjr+gA==}
     engines: {node: '>=18.0.0'}
@@ -542,8 +556,8 @@ packages:
   '@jridgewell/trace-mapping@0.3.31':
     resolution: {integrity: sha512-zzNR+SdQSDJzc8joaeP8QQoCQr8NuYx2dIIytl1QeBEZHJ9uW6hebsrYgbz8hJwUQao3TWCMtmfV8Nu1twOLAw==}
 
-  '@modelcontextprotocol/sdk@1.25.3':
-    resolution: {integrity: sha512-vsAMBMERybvYgKbg/l4L1rhS7VXV1c0CtyJg72vwxONVX0l4ZfKVAnZEWTQixJGTzKnELjQ59e4NbdFDALRiAQ==}
+  '@modelcontextprotocol/sdk@1.27.1':
+    resolution: {integrity: sha512-sr6GbP+4edBwFndLbM60gf07z0FQ79gaExpnsjMGePXqFcSSb7t6iscpjk9DhFhwd+mTEQrzNafGP8/iGGFYaA==}
     engines: {node: '>=18'}
     peerDependencies:
       '@cfworker/json-schema': ^4.1.1
@@ -552,6 +566,12 @@ packages:
       '@cfworker/json-schema':
         optional: true
 
+  '@napi-rs/wasm-runtime@1.1.2':
+    resolution: {integrity: sha512-sNXv5oLJ7ob93xkZ1XnxisYhGYXfaG9f65/ZgYuAu3qt7b3NadcOEhLvx28hv31PgX8SZJRYrAIPQilQmFpLVw==}
+    peerDependencies:
+      '@emnapi/core': ^1.7.1
+      '@emnapi/runtime': ^1.7.1
+
   '@octokit/auth-token@6.0.0':
     resolution: {integrity: sha512-P4YJBPdPSpWTQ1NU4XYdvHvXJJDxM6YwpS0FZHRgP7YFkdVxsWcpWGy/NVqlAA7PcPCnMacXlRm1y2PFZRWL/w==}
     engines: {node: '>= 20'}
@@ -600,6 +620,9 @@ packages:
   '@octokit/types@16.0.0':
     resolution: {integrity: sha512-sKq+9r1Mm4efXW1FCk7hFSeJo4QKreL/tTbR0rz/qx/r1Oa2VV83LTA/H/MuCOX7uCIJmQVRKBcbmWoySjAnSg==}
 
+  '@oxc-project/types@0.122.0':
+    resolution: {integrity: sha512-oLAl5kBpV4w69UtFZ9xqcmTi+GENWOcPF7FCrczTiBbmC0ibXxCwyvZGbO39rCVEuLGAZM84DH0pUIyyv/YJzA==}
+
   '@pnpm/config.env-replace@1.1.0':
     resolution: {integrity: sha512-htyl8TWnKL7K/ESFa1oW2UB5lVDxuF5DpM7tBi6Hu2LNL3mWkIzNLG6N4zoCUP1lCKNxWy/3iu8mS8MvToGd6w==}
     engines: {node: '>=12.22.0'}
@@ -612,128 +635,220 @@ packages:
     resolution: {integrity: sha512-h104Kh26rR8tm+a3Qkc5S4VLYint3FE48as7+/5oCEcKR2idC/pF1G6AhIXKI+eHPJa/3J9i5z0Al47IeGHPkA==}
     engines: {node: '>=12'}
 
-  '@rollup/rollup-android-arm-eabi@4.57.0':
-    resolution: {integrity: sha512-tPgXB6cDTndIe1ah7u6amCI1T0SsnlOuKgg10Xh3uizJk4e5M1JGaUMk7J4ciuAUcFpbOiNhm2XIjP9ON0dUqA==}
+  '@rolldown/binding-android-arm64@1.0.0-rc.12':
+    resolution: {integrity: sha512-pv1y2Fv0JybcykuiiD3qBOBdz6RteYojRFY1d+b95WVuzx211CRh+ytI/+9iVyWQ6koTh5dawe4S/yRfOFjgaA==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm64]
+    os: [android]
+
+  '@rolldown/binding-darwin-arm64@1.0.0-rc.12':
+    resolution: {integrity: sha512-cFYr6zTG/3PXXF3pUO+umXxt1wkRK/0AYT8lDwuqvRC+LuKYWSAQAQZjCWDQpAH172ZV6ieYrNnFzVVcnSflAg==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm64]
+    os: [darwin]
+
+  '@rolldown/binding-darwin-x64@1.0.0-rc.12':
+    resolution: {integrity: sha512-ZCsYknnHzeXYps0lGBz8JrF37GpE9bFVefrlmDrAQhOEi4IOIlcoU1+FwHEtyXGx2VkYAvhu7dyBf75EJQffBw==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [x64]
+    os: [darwin]
+
+  '@rolldown/binding-freebsd-x64@1.0.0-rc.12':
+    resolution: {integrity: sha512-dMLeprcVsyJsKolRXyoTH3NL6qtsT0Y2xeuEA8WQJquWFXkEC4bcu1rLZZSnZRMtAqwtrF/Ib9Ddtpa/Gkge9Q==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [x64]
+    os: [freebsd]
+
+  '@rolldown/binding-linux-arm-gnueabihf@1.0.0-rc.12':
+    resolution: {integrity: sha512-YqWjAgGC/9M1lz3GR1r1rP79nMgo3mQiiA+Hfo+pvKFK1fAJ1bCi0ZQVh8noOqNacuY1qIcfyVfP6HoyBRZ85Q==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm]
+    os: [linux]
+
+  '@rolldown/binding-linux-arm64-gnu@1.0.0-rc.12':
+    resolution: {integrity: sha512-/I5AS4cIroLpslsmzXfwbe5OmWvSsrFuEw3mwvbQ1kDxJ822hFHIx+vsN/TAzNVyepI/j/GSzrtCIwQPeKCLIg==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm64]
+    os: [linux]
+
+  '@rolldown/binding-linux-arm64-musl@1.0.0-rc.12':
+    resolution: {integrity: sha512-V6/wZztnBqlx5hJQqNWwFdxIKN0m38p8Jas+VoSfgH54HSj9tKTt1dZvG6JRHcjh6D7TvrJPWFGaY9UBVOaWPw==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm64]
+    os: [linux]
+
+  '@rolldown/binding-linux-ppc64-gnu@1.0.0-rc.12':
+    resolution: {integrity: sha512-AP3E9BpcUYliZCxa3w5Kwj9OtEVDYK6sVoUzy4vTOJsjPOgdaJZKFmN4oOlX0Wp0RPV2ETfmIra9x1xuayFB7g==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [ppc64]
+    os: [linux]
+
+  '@rolldown/binding-linux-s390x-gnu@1.0.0-rc.12':
+    resolution: {integrity: sha512-nWwpvUSPkoFmZo0kQazZYOrT7J5DGOJ/+QHHzjvNlooDZED8oH82Yg67HvehPPLAg5fUff7TfWFHQS8IV1n3og==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [s390x]
+    os: [linux]
+
+  '@rolldown/binding-linux-x64-gnu@1.0.0-rc.12':
+    resolution: {integrity: sha512-RNrafz5bcwRy+O9e6P8Z/OCAJW/A+qtBczIqVYwTs14pf4iV1/+eKEjdOUta93q2TsT/FI0XYDP3TCky38LMAg==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [x64]
+    os: [linux]
+
+  '@rolldown/binding-linux-x64-musl@1.0.0-rc.12':
+    resolution: {integrity: sha512-Jpw/0iwoKWx3LJ2rc1yjFrj+T7iHZn2JDg1Yny1ma0luviFS4mhAIcd1LFNxK3EYu3DHWCps0ydXQ5i/rrJ2ig==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [x64]
+    os: [linux]
+
+  '@rolldown/binding-openharmony-arm64@1.0.0-rc.12':
+    resolution: {integrity: sha512-vRugONE4yMfVn0+7lUKdKvN4D5YusEiPilaoO2sgUWpCvrncvWgPMzK00ZFFJuiPgLwgFNP5eSiUlv2tfc+lpA==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm64]
+    os: [openharmony]
+
+  '@rolldown/binding-wasm32-wasi@1.0.0-rc.12':
+    resolution: {integrity: sha512-ykGiLr/6kkiHc0XnBfmFJuCjr5ZYKKofkx+chJWDjitX+KsJuAmrzWhwyOMSHzPhzOHOy7u9HlFoa5MoAOJ/Zg==}
+    engines: {node: '>=14.0.0'}
+    cpu: [wasm32]
+
+  '@rolldown/binding-win32-arm64-msvc@1.0.0-rc.12':
+    resolution: {integrity: sha512-5eOND4duWkwx1AzCxadcOrNeighiLwMInEADT0YM7xeEOOFcovWZCq8dadXgcRHSf3Ulh1kFo/qvzoFiCLOL1Q==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [arm64]
+    os: [win32]
+
+  '@rolldown/binding-win32-x64-msvc@1.0.0-rc.12':
+    resolution: {integrity: sha512-PyqoipaswDLAZtot351MLhrlrh6lcZPo2LSYE+VDxbVk24LVKAGOuE4hb8xZQmrPAuEtTZW8E6D2zc5EUZX4Lw==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    cpu: [x64]
+    os: [win32]
+
+  '@rolldown/pluginutils@1.0.0-rc.12':
+    resolution: {integrity: sha512-HHMwmarRKvoFsJorqYlFeFRzXZqCt2ETQlEDOb9aqssrnVBB1/+xgTGtuTrIk5vzLNX1MjMtTf7W9z3tsSbrxw==}
+
+  '@rollup/rollup-android-arm-eabi@4.60.0':
+    resolution: {integrity: sha512-WOhNW9K8bR3kf4zLxbfg6Pxu2ybOUbB2AjMDHSQx86LIF4rH4Ft7vmMwNt0loO0eonglSNy4cpD3MKXXKQu0/A==}
     cpu: [arm]
     os: [android]
 
-  '@rollup/rollup-android-arm64@4.57.0':
-    resolution: {integrity: sha512-sa4LyseLLXr1onr97StkU1Nb7fWcg6niokTwEVNOO7awaKaoRObQ54+V/hrF/BP1noMEaaAW6Fg2d/CfLiq3Mg==}
+  '@rollup/rollup-android-arm64@4.60.0':
+    resolution: {integrity: sha512-u6JHLll5QKRvjciE78bQXDmqRqNs5M/3GVqZeMwvmjaNODJih/WIrJlFVEihvV0MiYFmd+ZyPr9wxOVbPAG2Iw==}
     cpu: [arm64]
     os: [android]
 
-  '@rollup/rollup-darwin-arm64@4.57.0':
-    resolution: {integrity: sha512-/NNIj9A7yLjKdmkx5dC2XQ9DmjIECpGpwHoGmA5E1AhU0fuICSqSWScPhN1yLCkEdkCwJIDu2xIeLPs60MNIVg==}
+  '@rollup/rollup-darwin-arm64@4.60.0':
+    resolution: {integrity: sha512-qEF7CsKKzSRc20Ciu2Zw1wRrBz4g56F7r/vRwY430UPp/nt1x21Q/fpJ9N5l47WWvJlkNCPJz3QRVw008fi7yA==}
     cpu: [arm64]
     os: [darwin]
 
-  '@rollup/rollup-darwin-x64@4.57.0':
-    resolution: {integrity: sha512-xoh8abqgPrPYPr7pTYipqnUi1V3em56JzE/HgDgitTqZBZ3yKCWI+7KUkceM6tNweyUKYru1UMi7FC060RyKwA==}
+  '@rollup/rollup-darwin-x64@4.60.0':
+    resolution: {integrity: sha512-WADYozJ4QCnXCH4wPB+3FuGmDPoFseVCUrANmA5LWwGmC6FL14BWC7pcq+FstOZv3baGX65tZ378uT6WG8ynTw==}
     cpu: [x64]
     os: [darwin]
 
-  '@rollup/rollup-freebsd-arm64@4.57.0':
-    resolution: {integrity: sha512-PCkMh7fNahWSbA0OTUQ2OpYHpjZZr0hPr8lId8twD7a7SeWrvT3xJVyza+dQwXSSq4yEQTMoXgNOfMCsn8584g==}
+  '@rollup/rollup-freebsd-arm64@4.60.0':
+    resolution: {integrity: sha512-6b8wGHJlDrGeSE3aH5mGNHBjA0TTkxdoNHik5EkvPHCt351XnigA4pS7Wsj/Eo9Y8RBU6f35cjN9SYmCFBtzxw==}
     cpu: [arm64]
     os: [freebsd]
 
-  '@rollup/rollup-freebsd-x64@4.57.0':
-    resolution: {integrity: sha512-1j3stGx+qbhXql4OCDZhnK7b01s6rBKNybfsX+TNrEe9JNq4DLi1yGiR1xW+nL+FNVvI4D02PUnl6gJ/2y6WJA==}
+  '@rollup/rollup-freebsd-x64@4.60.0':
+    resolution: {integrity: sha512-h25Ga0t4jaylMB8M/JKAyrvvfxGRjnPQIR8lnCayyzEjEOx2EJIlIiMbhpWxDRKGKF8jbNH01NnN663dH638mA==}
     cpu: [x64]
     os: [freebsd]
 
-  '@rollup/rollup-linux-arm-gnueabihf@4.57.0':
-    resolution: {integrity: sha512-eyrr5W08Ms9uM0mLcKfM/Uzx7hjhz2bcjv8P2uynfj0yU8GGPdz8iYrBPhiLOZqahoAMB8ZiolRZPbbU2MAi6Q==}
+  '@rollup/rollup-linux-arm-gnueabihf@4.60.0':
+    resolution: {integrity: sha512-RzeBwv0B3qtVBWtcuABtSuCzToo2IEAIQrcyB/b2zMvBWVbjo8bZDjACUpnaafaxhTw2W+imQbP2BD1usasK4g==}
     cpu: [arm]
     os: [linux]
 
-  '@rollup/rollup-linux-arm-musleabihf@4.57.0':
-    resolution: {integrity: sha512-Xds90ITXJCNyX9pDhqf85MKWUI4lqjiPAipJ8OLp8xqI2Ehk+TCVhF9rvOoN8xTbcafow3QOThkNnrM33uCFQA==}
+  '@rollup/rollup-linux-arm-musleabihf@4.60.0':
+    resolution: {integrity: sha512-Sf7zusNI2CIU1HLzuu9Tc5YGAHEZs5Lu7N1ssJG4Tkw6e0MEsN7NdjUDDfGNHy2IU+ENyWT+L2obgWiguWibWQ==}
     cpu: [arm]
     os: [linux]
 
-  '@rollup/rollup-linux-arm64-gnu@4.57.0':
-    resolution: {integrity: sha512-Xws2KA4CLvZmXjy46SQaXSejuKPhwVdaNinldoYfqruZBaJHqVo6hnRa8SDo9z7PBW5x84SH64+izmldCgbezw==}
+  '@rollup/rollup-linux-arm64-gnu@4.60.0':
+    resolution: {integrity: sha512-DX2x7CMcrJzsE91q7/O02IJQ5/aLkVtYFryqCjduJhUfGKG6yJV8hxaw8pZa93lLEpPTP/ohdN4wFz7yp/ry9A==}
     cpu: [arm64]
     os: [linux]
 
-  '@rollup/rollup-linux-arm64-musl@4.57.0':
-    resolution: {integrity: sha512-hrKXKbX5FdaRJj7lTMusmvKbhMJSGWJ+w++4KmjiDhpTgNlhYobMvKfDoIWecy4O60K6yA4SnztGuNTQF+Lplw==}
+  '@rollup/rollup-linux-arm64-musl@4.60.0':
+    resolution: {integrity: sha512-09EL+yFVbJZlhcQfShpswwRZ0Rg+z/CsSELFCnPt3iK+iqwGsI4zht3secj5vLEs957QvFFXnzAT0FFPIxSrkQ==}
     cpu: [arm64]
     os: [linux]
 
-  '@rollup/rollup-linux-loong64-gnu@4.57.0':
-    resolution: {integrity: sha512-6A+nccfSDGKsPm00d3xKcrsBcbqzCTAukjwWK6rbuAnB2bHaL3r9720HBVZ/no7+FhZLz/U3GwwZZEh6tOSI8Q==}
+  '@rollup/rollup-linux-loong64-gnu@4.60.0':
+    resolution: {integrity: sha512-i9IcCMPr3EXm8EQg5jnja0Zyc1iFxJjZWlb4wr7U2Wx/GrddOuEafxRdMPRYVaXjgbhvqalp6np07hN1w9kAKw==}
     cpu: [loong64]
     os: [linux]
 
-  '@rollup/rollup-linux-loong64-musl@4.57.0':
-    resolution: {integrity: sha512-4P1VyYUe6XAJtQH1Hh99THxr0GKMMwIXsRNOceLrJnaHTDgk1FTcTimDgneRJPvB3LqDQxUmroBclQ1S0cIJwQ==}
+  '@rollup/rollup-linux-loong64-musl@4.60.0':
+    resolution: {integrity: sha512-DGzdJK9kyJ+B78MCkWeGnpXJ91tK/iKA6HwHxF4TAlPIY7GXEvMe8hBFRgdrR9Ly4qebR/7gfUs9y2IoaVEyog==}
     cpu: [loong64]
     os: [linux]
 
-  '@rollup/rollup-linux-ppc64-gnu@4.57.0':
-    resolution: {integrity: sha512-8Vv6pLuIZCMcgXre6c3nOPhE0gjz1+nZP6T+hwWjr7sVH8k0jRkH+XnfjjOTglyMBdSKBPPz54/y1gToSKwrSQ==}
+  '@rollup/rollup-linux-ppc64-gnu@4.60.0':
+    resolution: {integrity: sha512-RwpnLsqC8qbS8z1H1AxBA1H6qknR4YpPR9w2XX0vo2Sz10miu57PkNcnHVaZkbqyw/kUWfKMI73jhmfi9BRMUQ==}
     cpu: [ppc64]
     os: [linux]
 
-  '@rollup/rollup-linux-ppc64-musl@4.57.0':
-    resolution: {integrity: sha512-r1te1M0Sm2TBVD/RxBPC6RZVwNqUTwJTA7w+C/IW5v9Ssu6xmxWEi+iJQlpBhtUiT1raJ5b48pI8tBvEjEFnFA==}
+  '@rollup/rollup-linux-ppc64-musl@4.60.0':
+    resolution: {integrity: sha512-Z8pPf54Ly3aqtdWC3G4rFigZgNvd+qJlOE52fmko3KST9SoGfAdSRCwyoyG05q1HrrAblLbk1/PSIV+80/pxLg==}
     cpu: [ppc64]
     os: [linux]
 
-  '@rollup/rollup-linux-riscv64-gnu@4.57.0':
-    resolution: {integrity: sha512-say0uMU/RaPm3CDQLxUUTF2oNWL8ysvHkAjcCzV2znxBr23kFfaxocS9qJm+NdkRhF8wtdEEAJuYcLPhSPbjuQ==}
+  '@rollup/rollup-linux-riscv64-gnu@4.60.0':
+    resolution: {integrity: sha512-3a3qQustp3COCGvnP4SvrMHnPQ9d1vzCakQVRTliaz8cIp/wULGjiGpbcqrkv0WrHTEp8bQD/B3HBjzujVWLOA==}
     cpu: [riscv64]
     os: [linux]
 
-  '@rollup/rollup-linux-riscv64-musl@4.57.0':
-    resolution: {integrity: sha512-/MU7/HizQGsnBREtRpcSbSV1zfkoxSTR7wLsRmBPQ8FwUj5sykrP1MyJTvsxP5KBq9SyE6kH8UQQQwa0ASeoQQ==}
+  '@rollup/rollup-linux-riscv64-musl@4.60.0':
+    resolution: {integrity: sha512-pjZDsVH/1VsghMJ2/kAaxt6dL0psT6ZexQVrijczOf+PeP2BUqTHYejk3l6TlPRydggINOeNRhvpLa0AYpCWSQ==}
     cpu: [riscv64]
     os: [linux]
 
-  '@rollup/rollup-linux-s390x-gnu@4.57.0':
-    resolution: {integrity: sha512-Q9eh+gUGILIHEaJf66aF6a414jQbDnn29zeu0eX3dHMuysnhTvsUvZTCAyZ6tJhUjnvzBKE4FtuaYxutxRZpOg==}
+  '@rollup/rollup-linux-s390x-gnu@4.60.0':
+    resolution: {integrity: sha512-3ObQs0BhvPgiUVZrN7gqCSvmFuMWvWvsjG5ayJ3Lraqv+2KhOsp+pUbigqbeWqueGIsnn+09HBw27rJ+gYK4VQ==}
     cpu: [s390x]
     os: [linux]
 
-  '@rollup/rollup-linux-x64-gnu@4.57.0':
-    resolution: {integrity: sha512-OR5p5yG5OKSxHReWmwvM0P+VTPMwoBS45PXTMYaskKQqybkS3Kmugq1W+YbNWArF8/s7jQScgzXUhArzEQ7x0A==}
+  '@rollup/rollup-linux-x64-gnu@4.60.0':
+    resolution: {integrity: sha512-EtylprDtQPdS5rXvAayrNDYoJhIz1/vzN2fEubo3yLE7tfAw+948dO0g4M0vkTVFhKojnF+n6C8bDNe+gDRdTg==}
     cpu: [x64]
     os: [linux]
 
-  '@rollup/rollup-linux-x64-musl@4.57.0':
-    resolution: {integrity: sha512-XeatKzo4lHDsVEbm1XDHZlhYZZSQYym6dg2X/Ko0kSFgio+KXLsxwJQprnR48GvdIKDOpqWqssC3iBCjoMcMpw==}
+  '@rollup/rollup-linux-x64-musl@4.60.0':
+    resolution: {integrity: sha512-k09oiRCi/bHU9UVFqD17r3eJR9bn03TyKraCrlz5ULFJGdJGi7VOmm9jl44vOJvRJ6P7WuBi/s2A97LxxHGIdw==}
     cpu: [x64]
     os: [linux]
 
-  '@rollup/rollup-openbsd-x64@4.57.0':
-    resolution: {integrity: sha512-Lu71y78F5qOfYmubYLHPcJm74GZLU6UJ4THkf/a1K7Tz2ycwC2VUbsqbJAXaR6Bx70SRdlVrt2+n5l7F0agTUw==}
+  '@rollup/rollup-openbsd-x64@4.60.0':
+    resolution: {integrity: sha512-1o/0/pIhozoSaDJoDcec+IVLbnRtQmHwPV730+AOD29lHEEo4F5BEUB24H0OBdhbBBDwIOSuf7vgg0Ywxdfiiw==}
     cpu: [x64]
     os: [openbsd]
 
-  '@rollup/rollup-openharmony-arm64@4.57.0':
-    resolution: {integrity: sha512-v5xwKDWcu7qhAEcsUubiav7r+48Uk/ENWdr82MBZZRIm7zThSxCIVDfb3ZeRRq9yqk+oIzMdDo6fCcA5DHfMyA==}
+  '@rollup/rollup-openharmony-arm64@4.60.0':
+    resolution: {integrity: sha512-pESDkos/PDzYwtyzB5p/UoNU/8fJo68vcXM9ZW2V0kjYayj1KaaUfi1NmTUTUpMn4UhU4gTuK8gIaFO4UGuMbA==}
     cpu: [arm64]
     os: [openharmony]
 
-  '@rollup/rollup-win32-arm64-msvc@4.57.0':
-    resolution: {integrity: sha512-XnaaaSMGSI6Wk8F4KK3QP7GfuuhjGchElsVerCplUuxRIzdvZ7hRBpLR0omCmw+kI2RFJB80nenhOoGXlJ5TfQ==}
+  '@rollup/rollup-win32-arm64-msvc@4.60.0':
+    resolution: {integrity: sha512-hj1wFStD7B1YBeYmvY+lWXZ7ey73YGPcViMShYikqKT1GtstIKQAtfUI6yrzPjAy/O7pO0VLXGmUVWXQMaYgTQ==}
     cpu: [arm64]
     os: [win32]
 
-  '@rollup/rollup-win32-ia32-msvc@4.57.0':
-    resolution: {integrity: sha512-3K1lP+3BXY4t4VihLw5MEg6IZD3ojSYzqzBG571W3kNQe4G4CcFpSUQVgurYgib5d+YaCjeFow8QivWp8vuSvA==}
+  '@rollup/rollup-win32-ia32-msvc@4.60.0':
+    resolution: {integrity: sha512-SyaIPFoxmUPlNDq5EHkTbiKzmSEmq/gOYFI/3HHJ8iS/v1mbugVa7dXUzcJGQfoytp9DJFLhHH4U3/eTy2Bq4w==}
     cpu: [ia32]
     os: [win32]
 
-  '@rollup/rollup-win32-x64-gnu@4.57.0':
-    resolution: {integrity: sha512-MDk610P/vJGc5L5ImE4k5s+GZT3en0KoK1MKPXCRgzmksAMk79j4h3k1IerxTNqwDLxsGxStEZVBqG0gIqZqoA==}
+  '@rollup/rollup-win32-x64-gnu@4.60.0':
+    resolution: {integrity: sha512-RdcryEfzZr+lAr5kRm2ucN9aVlCCa2QNq4hXelZxb8GG0NJSazq44Z3PCCc8wISRuCVnGs0lQJVX5Vp6fKA+IA==}
     cpu: [x64]
     os: [win32]
 
-  '@rollup/rollup-win32-x64-msvc@4.57.0':
-    resolution: {integrity: sha512-Zv7v6q6aV+VslnpwzqKAmrk5JdVkLUzok2208ZXGipjb+msxBr/fJPZyeEXiFgH7k62Ak0SLIfxQRZQvTuf7rQ==}
+  '@rollup/rollup-win32-x64-msvc@4.60.0':
+    resolution: {integrity: sha512-PrsWNQ8BuE00O3Xsx3ALh2Df8fAj9+cvvX9AIA6o4KpATR98c9mud4XtDWVvsEuyia5U4tVSTKygawyJkjm60w==}
     cpu: [x64]
     os: [win32]
 
@@ -801,6 +916,9 @@ packages:
   '@standard-schema/spec@1.1.0':
     resolution: {integrity: sha512-l2aFy5jALhniG5HgqrD6jXLi/rUWrKvqN/qJx6yoJsgKhblVd+iqqU4RCXavm/jPityDo5TCvKMnpjKnOriy0w==}
 
+  '@tybys/wasm-util@0.10.1':
+    resolution: {integrity: sha512-9tTaPJLSiejZKx+Bmog4uSubteqTvFrVrURwkmHixBo0G4seD0zUxp98E1DzUBJxLQ3NPwXrGKDiVjwx/DpPsg==}
+
   '@types/asciichart@1.5.8':
     resolution: {integrity: sha512-8yzgCUybv8/yUfj4WeTh7G+V+AxU7AzwsF2CrkTtARKHdrxE/EiByF2efUMj6qdm87tENucNu6pLs22RWU0H7g==}
 
@@ -906,7 +1024,7 @@ packages:
     resolution: {integrity: sha512-HhVd0MDnzzsgevnOWCBj5Otnzobjy5wLBe4EdeeFGv8luMsGcYqDuFRMcttKWZA5vVO8RFjexVovXvAM4JoJDQ==}
     peerDependencies:
       msw: ^2.4.9
-      vite: ^6.0.0 || ^7.0.0-0
+      vite: '>=7.3.2'
     peerDependenciesMeta:
       msw:
         optional: true
@@ -966,11 +1084,11 @@ packages:
       ajv:
         optional: true
 
-  ajv@6.12.6:
-    resolution: {integrity: sha512-j3fVLgvTo527anyYyJOGTYJbG+vnnQYvE0m5mmkc1TK+nxAppkCLMIL0aZ4dblVCNoGShhm+kzE4ZUykBoMg4g==}
+  ajv@6.14.0:
+    resolution: {integrity: sha512-IWrosm/yrn43eiKqkfkHis7QioDleaXQHdDVPKg0FSwwd/DuvyX79TZnFOnYpB7dcsFAMmtFztZuXPDvSePkFw==}
 
-  ajv@8.17.1:
-    resolution: {integrity: sha512-B/gBuNg5SiMTrPkC+A2+cW0RszwxYmn6VYxB/inlBStS5nx6xHIt/ehKRhIMhqusl7a8LjQoZnjCs5vhwxOQ1g==}
+  ajv@8.18.0:
+    resolution: {integrity: sha512-PlXPeEWMXMZ7sPYOHqmDyCJzcfNrUr3fGNKtezX14ykXOEIvyK81d+qydx89KY5O71FKMPaQ2vBfBFI5NHR63A==}
 
   ansi-align@3.0.1:
     resolution: {integrity: sha512-IOfwwBF5iczOjp/WeY4YxyjqAFMQoZufdQWDd19SEExbVLNXqvpzSJ/M7Za4/sCPmQ0+GRquoA7bGcINcxew6w==}
@@ -1021,8 +1139,9 @@ packages:
   ast-v8-to-istanbul@0.3.10:
     resolution: {integrity: sha512-p4K7vMz2ZSk3wN8l5o3y2bJAoZXT3VuJI5OLTATY/01CYWumWvwkUw0SqDBnNq6IiTO3qDa1eSQDibAV8g7XOQ==}
 
-  balanced-match@1.0.2:
-    resolution: {integrity: sha512-3oSeUO0TMV67hN1AmbXsK4yaqU7tjiHlbxRDZOpH0KW9+CeX4bRAaX0Anxt0tx2MrpRpWwQaPwIlISEJhYU5Pw==}
+  balanced-match@4.0.4:
+    resolution: {integrity: sha512-BLrgEcRTwX2o6gGxGOCNyMvGSp35YofuYzw9h1IMTRmKqttAZZVU67bdb9Pr2vUHA8+j3i2tJfjO6C6+4myGTA==}
+    engines: {node: 18 || 20 || >=22}
 
   before-after-hook@4.0.0:
     resolution: {integrity: sha512-q6tR3RPqIB1pMiTRMFcZwuG5T8vwp+vUvEG0vuI6B+Rikh5BfPp2fQ82c925FOs+b0lcFQ8CFrL+KbilfZFhOQ==}
@@ -1041,11 +1160,9 @@ packages:
     resolution: {integrity: sha512-F3PH5k5juxom4xktynS7MoFY+NUWH5LC4CnH11YB8NPew+HLpmBLCybSAEyb2F+4pRXhuhWqFesoQd6DAyc2hw==}
     engines: {node: '>=18'}
 
-  brace-expansion@1.1.12:
-    resolution: {integrity: sha512-9T9UjW3r0UW5c1Q7GTwllptXwhvYmEzFhzMfZ9H7FQWt+uZePjZPjBP/W1ZEyZ1twGWom5/56TF4lPcqjnDHcg==}
-
-  brace-expansion@2.0.2:
-    resolution: {integrity: sha512-Jt0vHyM+jmUBqojB7E1NIYadt0vI0Qxjxd2TErW94wDz+E2LAm5vKMXXwg6ZZBTHPuUlDgQHKXvjGBdfcF1ZDQ==}
+  brace-expansion@5.0.5:
+    resolution: {integrity: sha512-VZznLgtwhn+Mact9tfiwx64fA9erHH/MCXEUfB/0bX/6Fz6ny5EGTXYltMocqg4xFAQZtnO3DHWWXi8RiuN7cQ==}
+    engines: {node: 18 || 20 || >=22}
 
   braces@3.0.3:
     resolution: {integrity: sha512-yQbXgO/OSZVD2IsiLlro+7Hf6Q18EJrKSEsdoMzKePKXct3gvD8oLcOQdIzGupr5Fj+EDe8gO/lxc1BzfMpxvA==}
@@ -1176,9 +1293,6 @@ packages:
   compare-func@2.0.0:
     resolution: {integrity: sha512-zHig5N+tPWARooBnb0Zx1MFcdfpyJrfTJ3Y5L+IFvUm8rM74hHz66z0gw0x4tijh5CorKkKUCnW82R2vmpeCRA==}
 
-  concat-map@0.0.1:
-    resolution: {integrity: sha512-/Srv4dswyQNBfohGpz9o6Yb3Gz3SrUDqBH5rTuhGR7ahtlbYKnVxw2bCFMRljaA7EXHaXZ8wsHdodFvbkhKmqg==}
-
   confbox@0.1.8:
     resolution: {integrity: sha512-RMtmw0iFkeR4YV+fUOSucriAQNb9g8zFR52MWCtl+cCZOFRNL6zeB395vPzFhEjjn4fMxXudmELnl/KF/WrK6w==}
 
@@ -1303,6 +1417,10 @@ packages:
     resolution: {integrity: sha512-g7nH6P6dyDioJogAAGprGpCtVImJhpPk/roCzdb3fIh61/s/nPsfR6onyMwkCAR/OlC3yBC0lESvUoQEAssIrw==}
     engines: {node: '>= 0.8'}
 
+  detect-libc@2.1.2:
+    resolution: {integrity: sha512-Btj2BOOO83o3WyH59e8MgXsxEQVcarkUOpEYrubB0urwnN10yQ364rsiByU11nZlqWYZm05i/of7io4mzihBtQ==}
+    engines: {node: '>=8'}
+
   dir-glob@3.0.1:
     resolution: {integrity: sha512-WkrWp9GR4KXfKGYzOLmTuGVi1UWFfws377n9cc55/tb6DuqyF6pcQ5AbiHEshaDpY9v6oaSr2XCDidGmMwdzIA==}
     engines: {node: '>=8'}
@@ -1475,8 +1593,8 @@ packages:
     resolution: {integrity: sha512-knvyeauYhqjOYvQ66MznSMs83wmHrCycNEN6Ao+2AeYEfxUIkuiVxdEa1qlGEPK+We3n0THiDciYSsCcgW/DoA==}
     engines: {node: '>=12.0.0'}
 
-  express-rate-limit@7.5.1:
-    resolution: {integrity: sha512-7iN8iPMDzOMHPUYllBEsQdWVB6fPDMPqwjBaFrgr4Jgr/+okjvzAy+UHlYYL/Vs0OsOrMkwS6PJDkFlJwoxUnw==}
+  express-rate-limit@8.3.1:
+    resolution: {integrity: sha512-D1dKN+cmyPWuvB+G2SREQDzPY1agpBIcTa9sJxOPMCNeH3gwzhqJRDWCXW3gg0y//+LQ/8j52JbMROWyrKdMdw==}
     engines: {node: '>= 16'}
     peerDependencies:
       express: '>= 4.11'
@@ -1504,7 +1622,7 @@ packages:
     resolution: {integrity: sha512-tIbYtZbucOs0BRGqPJkshJUYdL+SDH7dVM8gjy+ERp3WAUjLEFJE+02kanyHtwjWOnwrKYBiwAmM0p4kLJAnXg==}
     engines: {node: '>=12.0.0'}
     peerDependencies:
-      picomatch: ^3 || ^4
+      picomatch: '>=4.0.4'
     peerDependenciesMeta:
       picomatch:
         optional: true
@@ -1561,8 +1679,8 @@ packages:
     resolution: {integrity: sha512-f7ccFPK3SXFHpx15UIGyRJ/FJQctuKZ0zVuN3frBo4HnK3cay9VEW0R6yPYFHC0AgqhukPzKjq22t5DmAyqGyw==}
     engines: {node: '>=16'}
 
-  flatted@3.3.3:
-    resolution: {integrity: sha512-GX+ysw4PBCz0PzosHDepZGANEuFCMLrnRTiEy9McGjmkCQYwRq4A/X786G/fjM/+OjsWSU1ZrY5qyARZmO/uwg==}
+  flatted@3.4.2:
+    resolution: {integrity: sha512-PjDse7RzhcPkIJwy5t7KPWQSZ9cAbzQXcafsetQoD7sOJRQlGikNbx7yZp2OotDnJyrDcbyRq3Ttb18iYOqkxA==}
 
   forwarded@0.2.0:
     resolution: {integrity: sha512-buRG0fpBtRHSTCOASe6hD258tEubFoRLb4ZNA6NxMVHNw2gOcwHo9wyablzMzOA5z9xA9L1KNjk/Nt6MT9aYow==}
@@ -1657,8 +1775,8 @@ packages:
   graceful-fs@4.2.11:
     resolution: {integrity: sha512-RbJ5/jmFcNNCcDV5o9eTnBLJ/HszWV0P73bc+Ff4nS/rJj+YaS6IGyiOL0VoBYX+l1Wrl3k63h/KrH+nhJ0XvQ==}
 
-  handlebars@4.7.8:
-    resolution: {integrity: sha512-vafaFqs8MZkRrSX7sFVUdo3ap/eNiLnb4IakshzvP56X5Nr1iGKAIqdX6tMlm6HcNRIkr6AxO5jFEoJzzpT8aQ==}
+  handlebars@4.7.9:
+    resolution: {integrity: sha512-4E71E0rpOaQuJR2A3xDZ+GM1HyWYv1clR58tC8emQNeQe3RH7MAzSbat+V0wG78LQBo6m6bzSG/L4pBuCsgnUQ==}
     engines: {node: '>=0.4.7'}
     hasBin: true
 
@@ -1681,8 +1799,8 @@ packages:
   highlight.js@10.7.3:
     resolution: {integrity: sha512-tzcUFauisWKNHaRkN4Wjl/ZA07gENAjFl3J/c480dprkGTg5EQstgaNFqBfUqCq54kZRIEcreTsAgF/m2quD7A==}
 
-  hono@4.11.7:
-    resolution: {integrity: sha512-l7qMiNee7t82bH3SeyUCt9UF15EVmaBvsppY2zQtrbIhl/yzBTny+YUxsVjSjQ6gaqaeVtZmGocom8TzBlA4Yw==}
+  hono@4.12.8:
+    resolution: {integrity: sha512-VJCEvtrezO1IAR+kqEYnxUOoStaQPGrCmX3j4wDTNOcD1uRPFpGlwQUIW8niPuvHXaTUxeOUl5MMDGrl+tmO9A==}
     engines: {node: '>=16.9.0'}
 
   hook-std@4.0.0:
@@ -1782,6 +1900,10 @@ packages:
     resolution: {integrity: sha512-2dYz766i9HprMBasCMvHMuazJ7u4WzhJwo5kb3iPSiW/iRYV6uPari3zHoqZlnuaR7V1bEiNMxikhp37rdBXbw==}
     engines: {node: '>=12'}
 
+  ip-address@10.1.0:
+    resolution: {integrity: sha512-XXADHxXmvT9+CRxhXg56LJovE+bmWnEWB78LB83VZTprKTmaC5QfruXocxzTZ2Kl0DNwKuBdlIhjL8LeY8Sf8Q==}
+    engines: {node: '>= 12'}
+
   ipaddr.js@1.9.1:
     resolution: {integrity: sha512-0KI/607xoxSToH7GjN1FfSbLoU0+btTicjsQSWQlh/hZykN8KpmMf7uYwPW3R+akZ6R/w18ZlXSHBYXiYUPO3g==}
     engines: {node: '>= 0.10'}
@@ -1933,6 +2055,76 @@ packages:
     resolution: {integrity: sha512-+bT2uH4E5LGE7h/n3evcS/sQlJXCpIp6ym8OWJ5eV6+67Dsql/LaaT7qJBAt2rzfoa/5QBGBhxDix1dMt2kQKQ==}
     engines: {node: '>= 0.8.0'}
 
+  lightningcss-android-arm64@1.32.0:
+    resolution: {integrity: sha512-YK7/ClTt4kAK0vo6w3X+Pnm0D2cf2vPHbhOXdoNti1Ga0al1P4TBZhwjATvjNwLEBCnKvjJc2jQgHXH0NEwlAg==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [arm64]
+    os: [android]
+
+  lightningcss-darwin-arm64@1.32.0:
+    resolution: {integrity: sha512-RzeG9Ju5bag2Bv1/lwlVJvBE3q6TtXskdZLLCyfg5pt+HLz9BqlICO7LZM7VHNTTn/5PRhHFBSjk5lc4cmscPQ==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [arm64]
+    os: [darwin]
+
+  lightningcss-darwin-x64@1.32.0:
+    resolution: {integrity: sha512-U+QsBp2m/s2wqpUYT/6wnlagdZbtZdndSmut/NJqlCcMLTWp5muCrID+K5UJ6jqD2BFshejCYXniPDbNh73V8w==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [x64]
+    os: [darwin]
+
+  lightningcss-freebsd-x64@1.32.0:
+    resolution: {integrity: sha512-JCTigedEksZk3tHTTthnMdVfGf61Fky8Ji2E4YjUTEQX14xiy/lTzXnu1vwiZe3bYe0q+SpsSH/CTeDXK6WHig==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [x64]
+    os: [freebsd]
+
+  lightningcss-linux-arm-gnueabihf@1.32.0:
+    resolution: {integrity: sha512-x6rnnpRa2GL0zQOkt6rts3YDPzduLpWvwAF6EMhXFVZXD4tPrBkEFqzGowzCsIWsPjqSK+tyNEODUBXeeVHSkw==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [arm]
+    os: [linux]
+
+  lightningcss-linux-arm64-gnu@1.32.0:
+    resolution: {integrity: sha512-0nnMyoyOLRJXfbMOilaSRcLH3Jw5z9HDNGfT/gwCPgaDjnx0i8w7vBzFLFR1f6CMLKF8gVbebmkUN3fa/kQJpQ==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [arm64]
+    os: [linux]
+
+  lightningcss-linux-arm64-musl@1.32.0:
+    resolution: {integrity: sha512-UpQkoenr4UJEzgVIYpI80lDFvRmPVg6oqboNHfoH4CQIfNA+HOrZ7Mo7KZP02dC6LjghPQJeBsvXhJod/wnIBg==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [arm64]
+    os: [linux]
+
+  lightningcss-linux-x64-gnu@1.32.0:
+    resolution: {integrity: sha512-V7Qr52IhZmdKPVr+Vtw8o+WLsQJYCTd8loIfpDaMRWGUZfBOYEJeyJIkqGIDMZPwPx24pUMfwSxxI8phr/MbOA==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [x64]
+    os: [linux]
+
+  lightningcss-linux-x64-musl@1.32.0:
+    resolution: {integrity: sha512-bYcLp+Vb0awsiXg/80uCRezCYHNg1/l3mt0gzHnWV9XP1W5sKa5/TCdGWaR/zBM2PeF/HbsQv/j2URNOiVuxWg==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [x64]
+    os: [linux]
+
+  lightningcss-win32-arm64-msvc@1.32.0:
+    resolution: {integrity: sha512-8SbC8BR40pS6baCM8sbtYDSwEVQd4JlFTOlaD3gWGHfThTcABnNDBda6eTZeqbofalIJhFx0qKzgHJmcPTnGdw==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [arm64]
+    os: [win32]
+
+  lightningcss-win32-x64-msvc@1.32.0:
+    resolution: {integrity: sha512-Amq9B/SoZYdDi1kFrojnoqPLxYhQ4Wo5XiL8EVJrVsB8ARoC1PWW6VGtT0WKCemjy8aC+louJnjS7U18x3b06Q==}
+    engines: {node: '>= 12.0.0'}
+    cpu: [x64]
+    os: [win32]
+
+  lightningcss@1.32.0:
+    resolution: {integrity: sha512-NXYBzinNrblfraPGyrbPoD19C1h9lfI/1mzgWYvXUTe414Gz/X1FD2XBZSZM7rRTrMA8JL3OtAaGifrIKhQ5yQ==}
+    engines: {node: '>= 12.0.0'}
+
   lilconfig@3.1.3:
     resolution: {integrity: sha512-/vlFKAoH5Cgt3Ie+JLhRbwOsCQePABiU3tJ1egGvyQ+33R/vcwM2Zl2QR/LzjsBeItPt3oSVXapn+m4nQDvpzw==}
     engines: {node: '>=14'}
@@ -1969,8 +2161,8 @@ packages:
     resolution: {integrity: sha512-gvVijfZvn7R+2qyPX8mAuKcFGDf6Nc61GdvGafQsHL0sBIxfKzA+usWn4GFC/bk+QdwPUD4kWFJLhElipq+0VA==}
     engines: {node: ^12.20.0 || ^14.13.1 || >=16.0.0}
 
-  lodash-es@4.17.23:
-    resolution: {integrity: sha512-kVI48u3PZr38HdYz98UmfPnXl2DXrpdctLrFLCd3kOx1xUkOmpFPx7gCWWM5MPkL/fD8zb+Ph0QzjGFs4+hHWg==}
+  lodash-es@4.18.1:
+    resolution: {integrity: sha512-J8xewKD/Gk22OZbhpOVSwcs60zhd95ESDwezOFuA3/099925PdHJ7OFHNTGtajL3AlZkykD32HykiMo+BIBI8A==}
 
   lodash.camelcase@4.3.0:
     resolution: {integrity: sha512-TwuEnCnxbc3rAvhf/LbG7tJUDzhqXyFnv3dtzLOPgCG/hODL7WFnsbwktkD7yUV0RrreP/l1PALq/YSg6VvjlA==}
@@ -2011,8 +2203,8 @@ packages:
   lodash.upperfirst@4.3.1:
     resolution: {integrity: sha512-sReKOYJIJf74dhJONhU4e0/shzi1trVbSWDOhKYE5XV2O+H7Sb2Dihwuc7xWxVl+DgFPyTqIN3zMfT9cq5iWDg==}
 
-  lodash@4.17.23:
-    resolution: {integrity: sha512-LgVTMpQtIopCi79SJeDiP0TfWi5CNEc/L/aRdTh3yIvmZXTnheWpKjSZhnvMl8iXbC1tFg9gdHHDMLoV7CnG+w==}
+  lodash@4.18.1:
+    resolution: {integrity: sha512-dMInicTPVE8d1e5otfwmmjlxkZoUpiVLwyeTdUsi/Caj/gfzzblBcCE5sRHV/AsjuCmxWrte2TNGSYuCeCq+0Q==}
 
   log-symbols@7.0.1:
     resolution: {integrity: sha512-ja1E3yCr9i/0hmBVaM0bfwDjnGy8I/s6PP4DFp+yP+a+mrHO4Rm7DtmnqROTUkHIkqffC84YY7AeqX6oFk0WFg==}
@@ -2106,16 +2298,9 @@ packages:
     resolution: {integrity: sha512-VP79XUPxV2CigYP3jWwAUFSku2aKqBH7uTAapFWCBqutsbmDo96KY5o8uh6U+/YSIn5OxJnXp73beVkpqMIGhA==}
     engines: {node: '>=18'}
 
-  minimatch@10.1.1:
-    resolution: {integrity: sha512-enIvLvRAFZYXJzkCYG5RKmPfrFArdLv+R+lbQ53BmIMLIry74bjKzX6iHAm8WYamJkhSSEabrWN5D97XnKObjQ==}
-    engines: {node: 20 || >=22}
-
-  minimatch@3.1.2:
-    resolution: {integrity: sha512-J7p63hRiAjw1NDEww1W7i37+ByIrOWO5XQQAzZ3VOcL0PNybwpfmV/N05zFAzwQ9USyEcX6t3UO+K5aqBQOIHw==}
-
-  minimatch@9.0.5:
-    resolution: {integrity: sha512-G6T0ZX48xgozx7587koeX9Ys2NYy6Gmv//P89sEte9V9whIapMNF4idKxnW2QtCcLiTWlb/wfCabAtAFWhhBow==}
-    engines: {node: '>=16 || 14 >=14.17'}
+  minimatch@10.2.4:
+    resolution: {integrity: sha512-oRjTw/97aTBN0RHbYCdtF1MQfvusSIBQM0IZEgzl6426+8jSC0nF1a/GmnVLpfB9yyr6g6FTqWqiZVbxrtaCIg==}
+    engines: {node: 18 || 20 || >=22}
 
   minimist@1.2.8:
     resolution: {integrity: sha512-2yyAR8qBkN3YuheJanUpWC5U3bb5osDywNB8RzDVlDwDHbocAJveqqj1u8+SVD7jkWT4yvsHCpWqqWqAxb0zCA==}
@@ -2456,8 +2641,8 @@ packages:
     resolution: {integrity: sha512-oWyT4gICAu+kaA7QWk/jvCHWarMKNs6pXOGWKDTr7cw4IGcUbW+PeTfbaQiLGheFRpjo6O9J0PmyMfQPjH71oA==}
     engines: {node: 20 || >=22}
 
-  path-to-regexp@8.3.0:
-    resolution: {integrity: sha512-7jdwVIRtsP8MYpdXSwOS0YdD0Du+qOoF/AEPIt88PcCFrZCzx41oxku1jD88hZBwbNUIEfpqvuhjFaMAqMTWnA==}
+  path-to-regexp@8.4.2:
+    resolution: {integrity: sha512-qRcuIdP69NPm4qbACK+aDogI5CBDMi1jKe0ry5rSQJz8JVLsC7jV8XpiJjGRLLol3N+R5ihGYcrPLTno6pAdBA==}
 
   path-type@4.0.0:
     resolution: {integrity: sha512-gDKb8aZMDeD/tZWs9P6+q0J9Mwkdl6xMV8TjnGP3qJVJ06bdMgkbBlLU8IdfOsIsFz2BW1rNVT3XuNEl8zPAvw==}
@@ -2469,12 +2654,8 @@ packages:
   picocolors@1.1.1:
     resolution: {integrity: sha512-xceH2snhtb5M9liqDsmEw56le376mTZkEX/jEb/RxNFyegNul7eNslCXP9FDj/Lcu0X8KEyMceP2ntpaHrDEVA==}
 
-  picomatch@2.3.1:
-    resolution: {integrity: sha512-JU3teHTNjmE2VCGFzuY8EXzCDVwEqB2a8fsIvwaStHhAWJEeVd1o1QD80CU6+ZdEXXSLbSsuLwJjkCBWqRQUVA==}
-    engines: {node: '>=8.6'}
-
-  picomatch@4.0.3:
-    resolution: {integrity: sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==}
+  picomatch@4.0.4:
+    resolution: {integrity: sha512-QP88BAKvMam/3NxH6vj2o21R6MjxZUAd6nlwAS/pnGvN9IVLocLHxGYIzFhg6fUQ+5th6P4dv4eW9jX3DSIj7A==}
     engines: {node: '>=12'}
 
   pidtree@0.6.0:
@@ -2508,7 +2689,7 @@ packages:
       jiti: '>=1.21.0'
       postcss: '>=8.0.9'
       tsx: ^4.8.1
-      yaml: ^2.4.2
+      yaml: '>=2.8.3'
     peerDependenciesMeta:
       jiti:
         optional: true
@@ -2519,8 +2700,8 @@ packages:
       yaml:
         optional: true
 
-  postcss@8.5.6:
-    resolution: {integrity: sha512-3Ybi1tAuwAP9s0r1UQ2J4n5Y0G05bJkpUIO0/bI9MhwmD70S5aTWbXGBwxHrelT+XM1k6dM0pk+SwNkpTRN7Pg==}
+  postcss@8.5.8:
+    resolution: {integrity: sha512-OW/rX8O/jXnm82Ey1k44pObPtdblfiuWnrd8X7GJ7emImCOstunGbXUpp7HdBrFQX6rJzn3sPT397Wp5aCwCHg==}
     engines: {node: ^10 || ^12 || >=14}
 
   prelude-ls@1.2.1:
@@ -2554,8 +2735,8 @@ packages:
     resolution: {integrity: sha512-vYt7UD1U9Wg6138shLtLOvdAu+8DsC/ilFtEVHcH+wydcSpNE20AfSOduf6MkRFahL5FY7X1oU7nKVZFtfq8Fg==}
     engines: {node: '>=6'}
 
-  qs@6.14.1:
-    resolution: {integrity: sha512-4EK3+xJl8Ts67nLYNwqw/dsFVnCf+qR7RgXSK9jEEm9unao3njwMDdmsdvoKBKHzxd7tCYz5e5M+SnMjdtXGQQ==}
+  qs@6.15.0:
+    resolution: {integrity: sha512-mAZTtNCeetKMH+pSjrb76NAM8V9a05I9aBZOHztWy/UqcJdQYNsf59vrRKWnojAT9Y+GbIvoTBC++CPHqpDBhQ==}
     engines: {node: '>=0.6'}
 
   range-parser@1.2.1:
@@ -2620,8 +2801,13 @@ packages:
   rfdc@1.4.1:
     resolution: {integrity: sha512-q1b3N5QkRUWUl7iyylaaj3kOpIT0N2i9MqIEQXP73GVsN9cw3fdx8X63cEmWhJGi2PPCF23Ijp7ktmd39rawIA==}
 
-  rollup@4.57.0:
-    resolution: {integrity: sha512-e5lPJi/aui4TO1LpAXIRLySmwXSE8k3b9zoGfd42p67wzxog4WHjiZF3M2uheQih4DGyc25QEV4yRBbpueNiUA==}
+  rolldown@1.0.0-rc.12:
+    resolution: {integrity: sha512-yP4USLIMYrwpPHEFB5JGH1uxhcslv6/hL0OyvTuY+3qlOSJvZ7ntYnoWpehBxufkgN0cvXxppuTu5hHa/zPh+A==}
+    engines: {node: ^20.19.0 || >=22.12.0}
+    hasBin: true
+
+  rollup@4.60.0:
+    resolution: {integrity: sha512-yqjxruMGBQJ2gG4HtjZtAfXArHomazDHoFwFFmZZl0r7Pdo7qCIXKqKHZc8yeoMgzJJ+pO6pEEHa+V7uzWlrAQ==}
     engines: {node: '>=18.0.0', npm: '>=8.0.0'}
     hasBin: true
 
@@ -2635,16 +2821,11 @@ packages:
   safer-buffer@2.1.2:
     resolution: {integrity: sha512-YZo3K82SD7Riyi0E1EQPojLz7kpepnSQI9IyPbHHg1XXXevb5dJI7tpyN2ADxGcQbHG7vcyRHk0cbwqcQriUtg==}
 
-  semantic-release@25.0.2:
-    resolution: {integrity: sha512-6qGjWccl5yoyugHt3jTgztJ9Y0JVzyH8/Voc/D8PlLat9pwxQYXz7W1Dpnq5h0/G5GCYGUaDSlYcyk3AMh5A6g==}
+  semantic-release@25.0.3:
+    resolution: {integrity: sha512-WRgl5GcypwramYX4HV+eQGzUbD7UUbljVmS+5G1uMwX/wLgYuJAxGeerXJDMO2xshng4+FXqCgyB5QfClV6WjA==}
     engines: {node: ^22.14.0 || >= 24.10.0}
     hasBin: true
 
-  semver-diff@5.0.0:
-    resolution: {integrity: sha512-0HbGtOm+S7T6NGQ/pxJSJipJvc4DK3FcRVMRkhsIwJDJ4Jcz5DQC1cPPzB5GhzyHjwttW878HaWQq46CkL3cqg==}
-    engines: {node: '>=12'}
-    deprecated: Deprecated as the semver package now supports this built-in.
-
   semver-regex@4.0.5:
     resolution: {integrity: sha512-hunMQrEy1T6Jr2uEVjrAIqjwWcQTgOAcIM52C8MY1EZSD3DDNft04XzvYKPqjED65bNVVko0YI38nYeEHCX3yw==}
     engines: {node: '>=12'}
@@ -2916,6 +3097,9 @@ packages:
   ts-interface-checker@0.1.13:
     resolution: {integrity: sha512-Y/arvbn+rrz3JCKl9C4kVNfTfSm2/mEp5FSz5EsZSANGPSlQrpRI5M4PKF+mJnE52jOO90PnPSc3Ur3bTQw0gA==}
 
+  tslib@2.8.1:
+    resolution: {integrity: sha512-oJFu94HQb+KVduSUQL7wnpmqnfmLsOA/nAh6b6EH0wCEoK0/mPeXU6c3wKDV83MkOuHPRHtSXKKU99IBazS/2w==}
+
   tsup@8.5.1:
     resolution: {integrity: sha512-xtgkqwdhpKWr3tKPmCkvYmS9xnQK3m3XgxZHwSUjvfTjp7YfXe5tT3GgWi0F2N+ZSMsOeWeZFh7ZZFg5iPhing==}
     engines: {node: '>=18'}
@@ -2986,8 +3170,8 @@ packages:
   undici-types@7.16.0:
     resolution: {integrity: sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw==}
 
-  undici@7.19.1:
-    resolution: {integrity: sha512-Gpq0iNm5M6cQWlyHQv9MV+uOj1jWk7LpkoE5vSp/7zjb4zMdAcUD+VL5y0nH4p9EbUklq00eVIIX/XcDHzu5xg==}
+  undici@7.24.5:
+    resolution: {integrity: sha512-3IWdCpjgxp15CbJnsi/Y9TCDE7HWVN19j1hmzVhoAkY/+CJx449tVxT5wZc1Gwg8J+P0LWvzlBzxYRnHJ+1i7Q==}
     engines: {node: '>=20.18.1'}
 
   unicode-emoji-modifier-base@1.0.0:
@@ -3034,31 +3218,34 @@ packages:
     resolution: {integrity: sha512-BNGbWLfd0eUPabhkXUVm0j8uuvREyTh5ovRa/dyow/BqAbZJyC+5fU+IzQOzmAKzYqYRAISoRhdQr3eIZ/PXqg==}
     engines: {node: '>= 0.8'}
 
-  vite@7.3.1:
-    resolution: {integrity: sha512-w+N7Hifpc3gRjZ63vYBXA56dvvRlNWRczTdmCBBa+CotUzAPf5b7YMdMR/8CQoeYE5LX3W4wj6RYTgonm1b9DA==}
+  vite@8.0.5:
+    resolution: {integrity: sha512-nmu43Qvq9UopTRfMx2jOYW5l16pb3iDC1JH6yMuPkpVbzK0k+L7dfsEDH4jRgYFmsg0sTAqkojoZgzLMlwHsCQ==}
     engines: {node: ^20.19.0 || >=22.12.0}
     hasBin: true
     peerDependencies:
       '@types/node': ^20.19.0 || >=22.12.0
+      '@vitejs/devtools': ^0.1.0
+      esbuild: ^0.27.0 || ^0.28.0
       jiti: '>=1.21.0'
       less: ^4.0.0
-      lightningcss: ^1.21.0
       sass: ^1.70.0
       sass-embedded: ^1.70.0
       stylus: '>=0.54.8'
       sugarss: ^5.0.0
       terser: ^5.16.0
       tsx: ^4.8.1
-      yaml: ^2.4.2
+      yaml: '>=2.8.3'
     peerDependenciesMeta:
       '@types/node':
         optional: true
+      '@vitejs/devtools':
+        optional: true
+      esbuild:
+        optional: true
       jiti:
         optional: true
       less:
         optional: true
-      lightningcss:
-        optional: true
       sass:
         optional: true
       sass-embedded:
@@ -3151,8 +3338,8 @@ packages:
     resolution: {integrity: sha512-0pfFzegeDWJHJIAmTLRP2DwHjdF5s7jo9tuztdQxAhINCdvS+3nGINqPd00AphqJR/0LhANUS6/+7SCb98YOfA==}
     engines: {node: '>=10'}
 
-  yaml@2.8.2:
-    resolution: {integrity: sha512-mplynKqc1C2hTVYxd0PU2xQAc22TI1vShAYGksCCfxbn/dFwnHTNi1bvYsBTkhdUNtGIf5xNOg938rrSSYvS9A==}
+  yaml@2.8.3:
+    resolution: {integrity: sha512-AvbaCLOO2Otw/lW5bmh9d/WEdcDFdQp2Z2ZUH3pX9U2ihyUY0nvLv7J6TrWowklRGPYbB/IuIMfYgxaCPg5Bpg==}
     engines: {node: '>= 14.6'}
     hasBin: true
 
@@ -3214,7 +3401,7 @@ snapshots:
   '@actions/http-client@3.0.2':
     dependencies:
       tunnel: 0.0.6
-      undici: 7.19.1
+      undici: 7.24.5
 
   '@actions/io@2.0.0': {}
 
@@ -3263,7 +3450,7 @@ snapshots:
   '@commitlint/config-validator@20.3.1':
     dependencies:
       '@commitlint/types': 20.3.1
-      ajv: 8.17.1
+      ajv: 8.18.0
 
   '@commitlint/ensure@20.3.1':
     dependencies:
@@ -3352,6 +3539,22 @@ snapshots:
       '@types/conventional-commits-parser': 5.0.2
       chalk: 5.6.2
 
+  '@emnapi/core@1.9.2':
+    dependencies:
+      '@emnapi/wasi-threads': 1.2.1
+      tslib: 2.8.1
+    optional: true
+
+  '@emnapi/runtime@1.9.2':
+    dependencies:
+      tslib: 2.8.1
+    optional: true
+
+  '@emnapi/wasi-threads@1.2.1':
+    dependencies:
+      tslib: 2.8.1
+    optional: true
+
   '@esbuild/aix-ppc64@0.27.2':
     optional: true
 
@@ -3441,7 +3644,7 @@ snapshots:
     dependencies:
       '@eslint/object-schema': 2.1.7
       debug: 4.4.3
-      minimatch: 3.1.2
+      minimatch: 10.2.4
     transitivePeerDependencies:
       - supports-color
 
@@ -3455,14 +3658,14 @@ snapshots:
 
   '@eslint/eslintrc@3.3.3':
     dependencies:
-      ajv: 6.12.6
+      ajv: 6.14.0
       debug: 4.4.3
       espree: 10.4.0
       globals: 14.0.0
       ignore: 5.3.2
       import-fresh: 3.3.1
       js-yaml: 4.1.1
-      minimatch: 3.1.2
+      minimatch: 10.2.4
       strip-json-comments: 3.1.1
     transitivePeerDependencies:
       - supports-color
@@ -3476,9 +3679,9 @@ snapshots:
       '@eslint/core': 0.17.0
       levn: 0.4.1
 
-  '@hono/node-server@1.19.9(hono@4.11.7)':
+  '@hono/node-server@1.19.11(hono@4.12.8)':
     dependencies:
-      hono: 4.11.7
+      hono: 4.12.8
 
   '@humanfs/core@0.19.1': {}
 
@@ -3491,12 +3694,6 @@ snapshots:
 
   '@humanwhocodes/retry@0.4.3': {}
 
-  '@isaacs/balanced-match@4.0.1': {}
-
-  '@isaacs/brace-expansion@5.0.0':
-    dependencies:
-      '@isaacs/balanced-match': 4.0.1
-
   '@jmlweb/commitlint-config@3.0.1(@commitlint/cli@20.3.1(@types/node@25.0.10)(typescript@5.9.3))(@commitlint/config-conventional@20.3.1)':
     dependencies:
       '@commitlint/cli': 20.3.1(@types/node@25.0.10)(typescript@5.9.3)
@@ -3524,13 +3721,13 @@ snapshots:
 
   '@jmlweb/tsconfig-base@1.0.5': {}
 
-  '@jmlweb/tsup-config-base@1.1.4(tsup@8.5.1(jiti@2.6.1)(postcss@8.5.6)(typescript@5.9.3)(yaml@2.8.2))':
+  '@jmlweb/tsup-config-base@1.1.4(tsup@8.5.1(jiti@2.6.1)(postcss@8.5.8)(typescript@5.9.3)(yaml@2.8.3))':
     dependencies:
-      tsup: 8.5.1(jiti@2.6.1)(postcss@8.5.6)(typescript@5.9.3)(yaml@2.8.2)
+      tsup: 8.5.1(jiti@2.6.1)(postcss@8.5.8)(typescript@5.9.3)(yaml@2.8.3)
 
-  '@jmlweb/vitest-config@2.0.0(vitest@4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2))':
+  '@jmlweb/vitest-config@2.0.0(vitest@4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3))':
     dependencies:
-      vitest: 4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2)
+      vitest: 4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3)
 
   '@jridgewell/gen-mapping@0.3.13':
     dependencies:
@@ -3546,18 +3743,19 @@ snapshots:
       '@jridgewell/resolve-uri': 3.1.2
       '@jridgewell/sourcemap-codec': 1.5.5
 
-  '@modelcontextprotocol/sdk@1.25.3(hono@4.11.7)(zod@4.3.6)':
+  '@modelcontextprotocol/sdk@1.27.1(zod@4.3.6)':
     dependencies:
-      '@hono/node-server': 1.19.9(hono@4.11.7)
-      ajv: 8.17.1
-      ajv-formats: 3.0.1(ajv@8.17.1)
+      '@hono/node-server': 1.19.11(hono@4.12.8)
+      ajv: 8.18.0
+      ajv-formats: 3.0.1(ajv@8.18.0)
       content-type: 1.0.5
       cors: 2.8.6
       cross-spawn: 7.0.6
       eventsource: 3.0.7
       eventsource-parser: 3.0.6
       express: 5.2.1
-      express-rate-limit: 7.5.1(express@5.2.1)
+      express-rate-limit: 8.3.1(express@5.2.1)
+      hono: 4.12.8
       jose: 6.1.3
       json-schema-typed: 8.0.2
       pkce-challenge: 5.0.1
@@ -3565,9 +3763,15 @@ snapshots:
       zod: 4.3.6
       zod-to-json-schema: 3.25.1(zod@4.3.6)
     transitivePeerDependencies:
-      - hono
       - supports-color
 
+  '@napi-rs/wasm-runtime@1.1.2(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)':
+    dependencies:
+      '@emnapi/core': 1.9.2
+      '@emnapi/runtime': 1.9.2
+      '@tybys/wasm-util': 0.10.1
+    optional: true
+
   '@octokit/auth-token@6.0.0': {}
 
   '@octokit/core@7.0.6':
@@ -3627,6 +3831,8 @@ snapshots:
     dependencies:
       '@octokit/openapi-types': 27.0.0
 
+  '@oxc-project/types@0.122.0': {}
+
   '@pnpm/config.env-replace@1.1.0': {}
 
   '@pnpm/network.ca-file@1.0.2':
@@ -3639,92 +3845,144 @@ snapshots:
       '@pnpm/network.ca-file': 1.0.2
       config-chain: 1.1.13
 
-  '@rollup/rollup-android-arm-eabi@4.57.0':
+  '@rolldown/binding-android-arm64@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-darwin-arm64@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-darwin-x64@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-freebsd-x64@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-arm-gnueabihf@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-arm64-gnu@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-arm64-musl@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-ppc64-gnu@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-s390x-gnu@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-x64-gnu@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-linux-x64-musl@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-openharmony-arm64@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-wasm32-wasi@1.0.0-rc.12(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)':
+    dependencies:
+      '@napi-rs/wasm-runtime': 1.1.2(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)
+    transitivePeerDependencies:
+      - '@emnapi/core'
+      - '@emnapi/runtime'
+    optional: true
+
+  '@rolldown/binding-win32-arm64-msvc@1.0.0-rc.12':
+    optional: true
+
+  '@rolldown/binding-win32-x64-msvc@1.0.0-rc.12':
     optional: true
 
-  '@rollup/rollup-android-arm64@4.57.0':
+  '@rolldown/pluginutils@1.0.0-rc.12': {}
+
+  '@rollup/rollup-android-arm-eabi@4.60.0':
     optional: true
 
-  '@rollup/rollup-darwin-arm64@4.57.0':
+  '@rollup/rollup-android-arm64@4.60.0':
     optional: true
 
-  '@rollup/rollup-darwin-x64@4.57.0':
+  '@rollup/rollup-darwin-arm64@4.60.0':
     optional: true
 
-  '@rollup/rollup-freebsd-arm64@4.57.0':
+  '@rollup/rollup-darwin-x64@4.60.0':
     optional: true
 
-  '@rollup/rollup-freebsd-x64@4.57.0':
+  '@rollup/rollup-freebsd-arm64@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-arm-gnueabihf@4.57.0':
+  '@rollup/rollup-freebsd-x64@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-arm-musleabihf@4.57.0':
+  '@rollup/rollup-linux-arm-gnueabihf@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-arm64-gnu@4.57.0':
+  '@rollup/rollup-linux-arm-musleabihf@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-arm64-musl@4.57.0':
+  '@rollup/rollup-linux-arm64-gnu@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-loong64-gnu@4.57.0':
+  '@rollup/rollup-linux-arm64-musl@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-loong64-musl@4.57.0':
+  '@rollup/rollup-linux-loong64-gnu@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-ppc64-gnu@4.57.0':
+  '@rollup/rollup-linux-loong64-musl@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-ppc64-musl@4.57.0':
+  '@rollup/rollup-linux-ppc64-gnu@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-riscv64-gnu@4.57.0':
+  '@rollup/rollup-linux-ppc64-musl@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-riscv64-musl@4.57.0':
+  '@rollup/rollup-linux-riscv64-gnu@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-s390x-gnu@4.57.0':
+  '@rollup/rollup-linux-riscv64-musl@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-x64-gnu@4.57.0':
+  '@rollup/rollup-linux-s390x-gnu@4.60.0':
     optional: true
 
-  '@rollup/rollup-linux-x64-musl@4.57.0':
+  '@rollup/rollup-linux-x64-gnu@4.60.0':
     optional: true
 
-  '@rollup/rollup-openbsd-x64@4.57.0':
+  '@rollup/rollup-linux-x64-musl@4.60.0':
     optional: true
 
-  '@rollup/rollup-openharmony-arm64@4.57.0':
+  '@rollup/rollup-openbsd-x64@4.60.0':
     optional: true
 
-  '@rollup/rollup-win32-arm64-msvc@4.57.0':
+  '@rollup/rollup-openharmony-arm64@4.60.0':
     optional: true
 
-  '@rollup/rollup-win32-ia32-msvc@4.57.0':
+  '@rollup/rollup-win32-arm64-msvc@4.60.0':
     optional: true
 
-  '@rollup/rollup-win32-x64-gnu@4.57.0':
+  '@rollup/rollup-win32-ia32-msvc@4.60.0':
     optional: true
 
-  '@rollup/rollup-win32-x64-msvc@4.57.0':
+  '@rollup/rollup-win32-x64-gnu@4.60.0':
+    optional: true
+
+  '@rollup/rollup-win32-x64-msvc@4.60.0':
     optional: true
 
   '@sec-ant/readable-stream@0.4.1': {}
 
-  '@semantic-release/changelog@6.0.3(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/changelog@6.0.3(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       '@semantic-release/error': 3.0.0
       aggregate-error: 3.1.0
       fs-extra: 11.3.3
-      lodash: 4.17.23
-      semantic-release: 25.0.2(typescript@5.9.3)
+      lodash: 4.18.1
+      semantic-release: 25.0.3(typescript@5.9.3)
 
-  '@semantic-release/commit-analyzer@13.0.1(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/commit-analyzer@13.0.1(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       conventional-changelog-angular: 8.1.0
       conventional-changelog-writer: 8.2.0
@@ -3732,9 +3990,9 @@ snapshots:
       conventional-commits-parser: 6.2.1
       debug: 4.4.3
       import-from-esm: 2.0.0
-      lodash-es: 4.17.23
+      lodash-es: 4.18.1
       micromatch: 4.0.8
-      semantic-release: 25.0.2(typescript@5.9.3)
+      semantic-release: 25.0.3(typescript@5.9.3)
     transitivePeerDependencies:
       - supports-color
 
@@ -3742,33 +4000,33 @@ snapshots:
 
   '@semantic-release/error@4.0.0': {}
 
-  '@semantic-release/exec@7.1.0(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/exec@7.1.0(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       '@semantic-release/error': 4.0.0
       aggregate-error: 3.1.0
       debug: 4.4.3
       execa: 9.6.1
-      lodash-es: 4.17.23
+      lodash-es: 4.18.1
       parse-json: 8.3.0
-      semantic-release: 25.0.2(typescript@5.9.3)
+      semantic-release: 25.0.3(typescript@5.9.3)
     transitivePeerDependencies:
       - supports-color
 
-  '@semantic-release/git@10.0.1(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/git@10.0.1(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       '@semantic-release/error': 3.0.0
       aggregate-error: 3.1.0
       debug: 4.4.3
       dir-glob: 3.0.1
       execa: 5.1.1
-      lodash: 4.17.23
+      lodash: 4.18.1
       micromatch: 4.0.8
       p-reduce: 2.1.0
-      semantic-release: 25.0.2(typescript@5.9.3)
+      semantic-release: 25.0.3(typescript@5.9.3)
     transitivePeerDependencies:
       - supports-color
 
-  '@semantic-release/github@12.0.2(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/github@12.0.2(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       '@octokit/core': 7.0.6
       '@octokit/plugin-paginate-rest': 14.0.0(@octokit/core@7.0.6)
@@ -3781,17 +4039,17 @@ snapshots:
       http-proxy-agent: 7.0.2
       https-proxy-agent: 7.0.6
       issue-parser: 7.0.1
-      lodash-es: 4.17.23
+      lodash-es: 4.18.1
       mime: 4.1.0
       p-filter: 4.1.0
-      semantic-release: 25.0.2(typescript@5.9.3)
+      semantic-release: 25.0.3(typescript@5.9.3)
       tinyglobby: 0.2.15
-      undici: 7.19.1
+      undici: 7.24.5
       url-join: 5.0.0
     transitivePeerDependencies:
       - supports-color
 
-  '@semantic-release/npm@13.1.3(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/npm@13.1.3(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       '@actions/core': 2.0.3
       '@semantic-release/error': 4.0.0
@@ -3799,18 +4057,18 @@ snapshots:
       env-ci: 11.2.0
       execa: 9.6.1
       fs-extra: 11.3.3
-      lodash-es: 4.17.23
+      lodash-es: 4.18.1
       nerf-dart: 1.0.0
       normalize-url: 8.1.1
       npm: 11.8.0
       rc: 1.2.8
       read-pkg: 10.0.0
       registry-auth-token: 5.1.1
-      semantic-release: 25.0.2(typescript@5.9.3)
+      semantic-release: 25.0.3(typescript@5.9.3)
       semver: 7.7.3
       tempy: 3.1.2
 
-  '@semantic-release/release-notes-generator@14.1.0(semantic-release@25.0.2(typescript@5.9.3))':
+  '@semantic-release/release-notes-generator@14.1.0(semantic-release@25.0.3(typescript@5.9.3))':
     dependencies:
       conventional-changelog-angular: 8.1.0
       conventional-changelog-writer: 8.2.0
@@ -3820,9 +4078,9 @@ snapshots:
       get-stream: 7.0.1
       import-from-esm: 2.0.0
       into-stream: 7.0.0
-      lodash-es: 4.17.23
+      lodash-es: 4.18.1
       read-package-up: 11.0.0
-      semantic-release: 25.0.2(typescript@5.9.3)
+      semantic-release: 25.0.3(typescript@5.9.3)
     transitivePeerDependencies:
       - supports-color
 
@@ -3832,6 +4090,11 @@ snapshots:
 
   '@standard-schema/spec@1.1.0': {}
 
+  '@tybys/wasm-util@0.10.1':
+    dependencies:
+      tslib: 2.8.1
+    optional: true
+
   '@types/asciichart@1.5.8': {}
 
   '@types/chai@5.2.3':
@@ -3929,7 +4192,7 @@ snapshots:
       '@typescript-eslint/types': 8.54.0
       '@typescript-eslint/visitor-keys': 8.54.0
       debug: 4.4.3
-      minimatch: 9.0.5
+      minimatch: 10.2.4
       semver: 7.7.3
       tinyglobby: 0.2.15
       ts-api-utils: 2.4.0(typescript@5.9.3)
@@ -3953,7 +4216,7 @@ snapshots:
       '@typescript-eslint/types': 8.54.0
       eslint-visitor-keys: 4.2.1
 
-  '@vitest/coverage-v8@4.0.18(vitest@4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2))':
+  '@vitest/coverage-v8@4.0.18(vitest@4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3))':
     dependencies:
       '@bcoe/v8-coverage': 1.0.2
       '@vitest/utils': 4.0.18
@@ -3965,7 +4228,7 @@ snapshots:
       obug: 2.1.1
       std-env: 3.10.0
       tinyrainbow: 3.0.3
-      vitest: 4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2)
+      vitest: 4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3)
 
   '@vitest/expect@4.0.18':
     dependencies:
@@ -3976,13 +4239,13 @@ snapshots:
       chai: 6.2.2
       tinyrainbow: 3.0.3
 
-  '@vitest/mocker@4.0.18(vite@7.3.1(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2))':
+  '@vitest/mocker@4.0.18(vite@8.0.5(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3))':
     dependencies:
       '@vitest/spy': 4.0.18
       estree-walker: 3.0.3
       magic-string: 0.30.21
     optionalDependencies:
-      vite: 7.3.1(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2)
+      vite: 8.0.5(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3)
 
   '@vitest/pretty-format@4.0.18':
     dependencies:
@@ -4034,18 +4297,18 @@ snapshots:
       clean-stack: 5.3.0
       indent-string: 5.0.0
 
-  ajv-formats@3.0.1(ajv@8.17.1):
+  ajv-formats@3.0.1(ajv@8.18.0):
     optionalDependencies:
-      ajv: 8.17.1
+      ajv: 8.18.0
 
-  ajv@6.12.6:
+  ajv@6.14.0:
     dependencies:
       fast-deep-equal: 3.1.3
       fast-json-stable-stringify: 2.1.0
       json-schema-traverse: 0.4.1
       uri-js: 4.4.1
 
-  ajv@8.17.1:
+  ajv@8.18.0:
     dependencies:
       fast-deep-equal: 3.1.3
       fast-uri: 3.1.0
@@ -4092,7 +4355,7 @@ snapshots:
       estree-walker: 3.0.3
       js-tokens: 9.0.1
 
-  balanced-match@1.0.2: {}
+  balanced-match@4.0.4: {}
 
   before-after-hook@4.0.0: {}
 
@@ -4106,7 +4369,7 @@ snapshots:
       http-errors: 2.0.1
       iconv-lite: 0.7.2
       on-finished: 2.4.1
-      qs: 6.14.1
+      qs: 6.15.0
       raw-body: 3.0.2
       type-is: 2.0.1
     transitivePeerDependencies:
@@ -4125,14 +4388,9 @@ snapshots:
       widest-line: 5.0.0
       wrap-ansi: 9.0.2
 
-  brace-expansion@1.1.12:
-    dependencies:
-      balanced-match: 1.0.2
-      concat-map: 0.0.1
-
-  brace-expansion@2.0.2:
+  brace-expansion@5.0.5:
     dependencies:
-      balanced-match: 1.0.2
+      balanced-match: 4.0.4
 
   braces@3.0.3:
     dependencies:
@@ -4257,8 +4515,6 @@ snapshots:
       array-ify: 1.0.0
       dot-prop: 5.3.0
 
-  concat-map@0.0.1: {}
-
   confbox@0.1.8: {}
 
   config-chain@1.1.13:
@@ -4291,7 +4547,7 @@ snapshots:
   conventional-changelog-writer@8.2.0:
     dependencies:
       conventional-commits-filter: 5.0.0
-      handlebars: 4.7.8
+      handlebars: 4.7.9
       meow: 13.2.0
       semver: 7.7.3
 
@@ -4361,6 +4617,8 @@ snapshots:
 
   depd@2.0.0: {}
 
+  detect-libc@2.1.2: {}
+
   dir-glob@3.0.1:
     dependencies:
       path-type: 4.0.0
@@ -4482,7 +4740,7 @@ snapshots:
       '@humanwhocodes/module-importer': 1.0.1
       '@humanwhocodes/retry': 0.4.3
       '@types/estree': 1.0.8
-      ajv: 6.12.6
+      ajv: 6.14.0
       chalk: 4.1.2
       cross-spawn: 7.0.6
       debug: 4.4.3
@@ -4501,7 +4759,7 @@ snapshots:
       is-glob: 4.0.3
       json-stable-stringify-without-jsonify: 1.0.1
       lodash.merge: 4.6.2
-      minimatch: 3.1.2
+      minimatch: 10.2.4
       natural-compare: 1.4.0
       optionator: 0.9.4
     optionalDependencies:
@@ -4582,9 +4840,10 @@ snapshots:
 
   expect-type@1.3.0: {}
 
-  express-rate-limit@7.5.1(express@5.2.1):
+  express-rate-limit@8.3.1(express@5.2.1):
     dependencies:
       express: 5.2.1
+      ip-address: 10.1.0
 
   express@5.2.1:
     dependencies:
@@ -4608,7 +4867,7 @@ snapshots:
       once: 1.4.0
       parseurl: 1.3.3
       proxy-addr: 2.0.7
-      qs: 6.14.1
+      qs: 6.15.0
       range-parser: 1.2.1
       router: 2.2.0
       send: 1.2.1
@@ -4629,9 +4888,9 @@ snapshots:
 
   fast-uri@3.1.0: {}
 
-  fdir@6.5.0(picomatch@4.0.3):
+  fdir@6.5.0(picomatch@4.0.4):
     optionalDependencies:
-      picomatch: 4.0.3
+      picomatch: 4.0.4
 
   figlet@1.10.0:
     dependencies:
@@ -4690,14 +4949,14 @@ snapshots:
     dependencies:
       magic-string: 0.30.21
       mlly: 1.8.0
-      rollup: 4.57.0
+      rollup: 4.60.0
 
   flat-cache@4.0.1:
     dependencies:
-      flatted: 3.3.3
+      flatted: 3.4.2
       keyv: 4.5.4
 
-  flatted@3.3.3: {}
+  flatted@3.4.2: {}
 
   forwarded@0.2.0: {}
 
@@ -4775,7 +5034,7 @@ snapshots:
 
   glob@13.0.0:
     dependencies:
-      minimatch: 10.1.1
+      minimatch: 10.2.4
       minipass: 7.1.2
       path-scurry: 2.0.1
 
@@ -4791,7 +5050,7 @@ snapshots:
 
   graceful-fs@4.2.11: {}
 
-  handlebars@4.7.8:
+  handlebars@4.7.9:
     dependencies:
       minimist: 1.2.8
       neo-async: 2.6.2
@@ -4812,7 +5071,7 @@ snapshots:
 
   highlight.js@10.7.3: {}
 
-  hono@4.11.7: {}
+  hono@4.12.8: {}
 
   hook-std@4.0.0: {}
 
@@ -4897,6 +5156,8 @@ snapshots:
       from2: 2.3.0
       p-is-promise: 3.0.0
 
+  ip-address@10.1.0: {}
+
   ipaddr.js@1.9.1: {}
 
   is-any-array@2.0.1: {}
@@ -5011,6 +5272,55 @@ snapshots:
       prelude-ls: 1.2.1
       type-check: 0.4.0
 
+  lightningcss-android-arm64@1.32.0:
+    optional: true
+
+  lightningcss-darwin-arm64@1.32.0:
+    optional: true
+
+  lightningcss-darwin-x64@1.32.0:
+    optional: true
+
+  lightningcss-freebsd-x64@1.32.0:
+    optional: true
+
+  lightningcss-linux-arm-gnueabihf@1.32.0:
+    optional: true
+
+  lightningcss-linux-arm64-gnu@1.32.0:
+    optional: true
+
+  lightningcss-linux-arm64-musl@1.32.0:
+    optional: true
+
+  lightningcss-linux-x64-gnu@1.32.0:
+    optional: true
+
+  lightningcss-linux-x64-musl@1.32.0:
+    optional: true
+
+  lightningcss-win32-arm64-msvc@1.32.0:
+    optional: true
+
+  lightningcss-win32-x64-msvc@1.32.0:
+    optional: true
+
+  lightningcss@1.32.0:
+    dependencies:
+      detect-libc: 2.1.2
+    optionalDependencies:
+      lightningcss-android-arm64: 1.32.0
+      lightningcss-darwin-arm64: 1.32.0
+      lightningcss-darwin-x64: 1.32.0
+      lightningcss-freebsd-x64: 1.32.0
+      lightningcss-linux-arm-gnueabihf: 1.32.0
+      lightningcss-linux-arm64-gnu: 1.32.0
+      lightningcss-linux-arm64-musl: 1.32.0
+      lightningcss-linux-x64-gnu: 1.32.0
+      lightningcss-linux-x64-musl: 1.32.0
+      lightningcss-win32-arm64-msvc: 1.32.0
+      lightningcss-win32-x64-msvc: 1.32.0
+
   lilconfig@3.1.3: {}
 
   lines-and-columns@1.2.4: {}
@@ -5023,7 +5333,7 @@ snapshots:
       nano-spawn: 2.0.0
       pidtree: 0.6.0
       string-argv: 0.3.2
-      yaml: 2.8.2
+      yaml: 2.8.3
 
   listr2@9.0.5:
     dependencies:
@@ -5056,7 +5366,7 @@ snapshots:
     dependencies:
       p-locate: 6.0.0
 
-  lodash-es@4.17.23: {}
+  lodash-es@4.18.1: {}
 
   lodash.camelcase@4.3.0: {}
 
@@ -5084,7 +5394,7 @@ snapshots:
 
   lodash.upperfirst@4.3.1: {}
 
-  lodash@4.17.23: {}
+  lodash@4.18.1: {}
 
   log-symbols@7.0.1:
     dependencies:
@@ -5151,7 +5461,7 @@ snapshots:
   micromatch@4.0.8:
     dependencies:
       braces: 3.0.3
-      picomatch: 2.3.1
+      picomatch: 4.0.4
 
   mime-db@1.54.0: {}
 
@@ -5167,17 +5477,9 @@ snapshots:
 
   mimic-function@5.0.1: {}
 
-  minimatch@10.1.1:
-    dependencies:
-      '@isaacs/brace-expansion': 5.0.0
-
-  minimatch@3.1.2:
-    dependencies:
-      brace-expansion: 1.1.12
-
-  minimatch@9.0.5:
+  minimatch@10.2.4:
     dependencies:
-      brace-expansion: 2.0.2
+      brace-expansion: 5.0.5
 
   minimist@1.2.8: {}
 
@@ -5448,7 +5750,7 @@ snapshots:
       lru-cache: 11.2.5
       minipass: 7.1.2
 
-  path-to-regexp@8.3.0: {}
+  path-to-regexp@8.4.2: {}
 
   path-type@4.0.0: {}
 
@@ -5456,9 +5758,7 @@ snapshots:
 
   picocolors@1.1.1: {}
 
-  picomatch@2.3.1: {}
-
-  picomatch@4.0.3: {}
+  picomatch@4.0.4: {}
 
   pidtree@0.6.0: {}
 
@@ -5479,15 +5779,15 @@ snapshots:
       mlly: 1.8.0
       pathe: 2.0.3
 
-  postcss-load-config@6.0.1(jiti@2.6.1)(postcss@8.5.6)(yaml@2.8.2):
+  postcss-load-config@6.0.1(jiti@2.6.1)(postcss@8.5.8)(yaml@2.8.3):
     dependencies:
       lilconfig: 3.1.3
     optionalDependencies:
       jiti: 2.6.1
-      postcss: 8.5.6
-      yaml: 2.8.2
+      postcss: 8.5.8
+      yaml: 2.8.3
 
-  postcss@8.5.6:
+  postcss@8.5.8:
     dependencies:
       nanoid: 3.3.11
       picocolors: 1.1.1
@@ -5517,7 +5817,7 @@ snapshots:
 
   punycode@2.3.1: {}
 
-  qs@6.14.1:
+  qs@6.15.0:
     dependencies:
       side-channel: 1.1.0
 
@@ -5596,35 +5896,59 @@ snapshots:
 
   rfdc@1.4.1: {}
 
-  rollup@4.57.0:
+  rolldown@1.0.0-rc.12(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2):
+    dependencies:
+      '@oxc-project/types': 0.122.0
+      '@rolldown/pluginutils': 1.0.0-rc.12
+    optionalDependencies:
+      '@rolldown/binding-android-arm64': 1.0.0-rc.12
+      '@rolldown/binding-darwin-arm64': 1.0.0-rc.12
+      '@rolldown/binding-darwin-x64': 1.0.0-rc.12
+      '@rolldown/binding-freebsd-x64': 1.0.0-rc.12
+      '@rolldown/binding-linux-arm-gnueabihf': 1.0.0-rc.12
+      '@rolldown/binding-linux-arm64-gnu': 1.0.0-rc.12
+      '@rolldown/binding-linux-arm64-musl': 1.0.0-rc.12
+      '@rolldown/binding-linux-ppc64-gnu': 1.0.0-rc.12
+      '@rolldown/binding-linux-s390x-gnu': 1.0.0-rc.12
+      '@rolldown/binding-linux-x64-gnu': 1.0.0-rc.12
+      '@rolldown/binding-linux-x64-musl': 1.0.0-rc.12
+      '@rolldown/binding-openharmony-arm64': 1.0.0-rc.12
+      '@rolldown/binding-wasm32-wasi': 1.0.0-rc.12(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)
+      '@rolldown/binding-win32-arm64-msvc': 1.0.0-rc.12
+      '@rolldown/binding-win32-x64-msvc': 1.0.0-rc.12
+    transitivePeerDependencies:
+      - '@emnapi/core'
+      - '@emnapi/runtime'
+
+  rollup@4.60.0:
     dependencies:
       '@types/estree': 1.0.8
     optionalDependencies:
-      '@rollup/rollup-android-arm-eabi': 4.57.0
-      '@rollup/rollup-android-arm64': 4.57.0
-      '@rollup/rollup-darwin-arm64': 4.57.0
-      '@rollup/rollup-darwin-x64': 4.57.0
-      '@rollup/rollup-freebsd-arm64': 4.57.0
-      '@rollup/rollup-freebsd-x64': 4.57.0
-      '@rollup/rollup-linux-arm-gnueabihf': 4.57.0
-      '@rollup/rollup-linux-arm-musleabihf': 4.57.0
-      '@rollup/rollup-linux-arm64-gnu': 4.57.0
-      '@rollup/rollup-linux-arm64-musl': 4.57.0
-      '@rollup/rollup-linux-loong64-gnu': 4.57.0
-      '@rollup/rollup-linux-loong64-musl': 4.57.0
-      '@rollup/rollup-linux-ppc64-gnu': 4.57.0
-      '@rollup/rollup-linux-ppc64-musl': 4.57.0
-      '@rollup/rollup-linux-riscv64-gnu': 4.57.0
-      '@rollup/rollup-linux-riscv64-musl': 4.57.0
-      '@rollup/rollup-linux-s390x-gnu': 4.57.0
-      '@rollup/rollup-linux-x64-gnu': 4.57.0
-      '@rollup/rollup-linux-x64-musl': 4.57.0
-      '@rollup/rollup-openbsd-x64': 4.57.0
-      '@rollup/rollup-openharmony-arm64': 4.57.0
-      '@rollup/rollup-win32-arm64-msvc': 4.57.0
-      '@rollup/rollup-win32-ia32-msvc': 4.57.0
-      '@rollup/rollup-win32-x64-gnu': 4.57.0
-      '@rollup/rollup-win32-x64-msvc': 4.57.0
+      '@rollup/rollup-android-arm-eabi': 4.60.0
+      '@rollup/rollup-android-arm64': 4.60.0
+      '@rollup/rollup-darwin-arm64': 4.60.0
+      '@rollup/rollup-darwin-x64': 4.60.0
+      '@rollup/rollup-freebsd-arm64': 4.60.0
+      '@rollup/rollup-freebsd-x64': 4.60.0
+      '@rollup/rollup-linux-arm-gnueabihf': 4.60.0
+      '@rollup/rollup-linux-arm-musleabihf': 4.60.0
+      '@rollup/rollup-linux-arm64-gnu': 4.60.0
+      '@rollup/rollup-linux-arm64-musl': 4.60.0
+      '@rollup/rollup-linux-loong64-gnu': 4.60.0
+      '@rollup/rollup-linux-loong64-musl': 4.60.0
+      '@rollup/rollup-linux-ppc64-gnu': 4.60.0
+      '@rollup/rollup-linux-ppc64-musl': 4.60.0
+      '@rollup/rollup-linux-riscv64-gnu': 4.60.0
+      '@rollup/rollup-linux-riscv64-musl': 4.60.0
+      '@rollup/rollup-linux-s390x-gnu': 4.60.0
+      '@rollup/rollup-linux-x64-gnu': 4.60.0
+      '@rollup/rollup-linux-x64-musl': 4.60.0
+      '@rollup/rollup-openbsd-x64': 4.60.0
+      '@rollup/rollup-openharmony-arm64': 4.60.0
+      '@rollup/rollup-win32-arm64-msvc': 4.60.0
+      '@rollup/rollup-win32-ia32-msvc': 4.60.0
+      '@rollup/rollup-win32-x64-gnu': 4.60.0
+      '@rollup/rollup-win32-x64-msvc': 4.60.0
       fsevents: 2.3.3
 
   router@2.2.0:
@@ -5633,7 +5957,7 @@ snapshots:
       depd: 2.0.0
       is-promise: 4.0.0
       parseurl: 1.3.3
-      path-to-regexp: 8.3.0
+      path-to-regexp: 8.4.2
     transitivePeerDependencies:
       - supports-color
 
@@ -5641,13 +5965,13 @@ snapshots:
 
   safer-buffer@2.1.2: {}
 
-  semantic-release@25.0.2(typescript@5.9.3):
+  semantic-release@25.0.3(typescript@5.9.3):
     dependencies:
-      '@semantic-release/commit-analyzer': 13.0.1(semantic-release@25.0.2(typescript@5.9.3))
+      '@semantic-release/commit-analyzer': 13.0.1(semantic-release@25.0.3(typescript@5.9.3))
       '@semantic-release/error': 4.0.0
-      '@semantic-release/github': 12.0.2(semantic-release@25.0.2(typescript@5.9.3))
-      '@semantic-release/npm': 13.1.3(semantic-release@25.0.2(typescript@5.9.3))
-      '@semantic-release/release-notes-generator': 14.1.0(semantic-release@25.0.2(typescript@5.9.3))
+      '@semantic-release/github': 12.0.2(semantic-release@25.0.3(typescript@5.9.3))
+      '@semantic-release/npm': 13.1.3(semantic-release@25.0.3(typescript@5.9.3))
+      '@semantic-release/release-notes-generator': 14.1.0(semantic-release@25.0.3(typescript@5.9.3))
       aggregate-error: 5.0.0
       cosmiconfig: 9.0.0(typescript@5.9.3)
       debug: 4.4.3
@@ -5660,7 +5984,7 @@ snapshots:
       hook-std: 4.0.0
       hosted-git-info: 9.0.2
       import-from-esm: 2.0.0
-      lodash-es: 4.17.23
+      lodash-es: 4.18.1
       marked: 15.0.12
       marked-terminal: 7.3.0(marked@15.0.12)
       micromatch: 4.0.8
@@ -5669,17 +5993,12 @@ snapshots:
       read-package-up: 12.0.0
       resolve-from: 5.0.0
       semver: 7.7.3
-      semver-diff: 5.0.0
       signale: 1.4.0
       yargs: 18.0.0
     transitivePeerDependencies:
       - supports-color
       - typescript
 
-  semver-diff@5.0.0:
-    dependencies:
-      semver: 7.7.3
-
   semver-regex@4.0.5: {}
 
   semver@7.7.3: {}
@@ -5923,8 +6242,8 @@ snapshots:
 
   tinyglobby@0.2.15:
     dependencies:
-      fdir: 6.5.0(picomatch@4.0.3)
-      picomatch: 4.0.3
+      fdir: 6.5.0(picomatch@4.0.4)
+      picomatch: 4.0.4
 
   tinyrainbow@3.0.3: {}
 
@@ -5944,7 +6263,10 @@ snapshots:
 
   ts-interface-checker@0.1.13: {}
 
-  tsup@8.5.1(jiti@2.6.1)(postcss@8.5.6)(typescript@5.9.3)(yaml@2.8.2):
+  tslib@2.8.1:
+    optional: true
+
+  tsup@8.5.1(jiti@2.6.1)(postcss@8.5.8)(typescript@5.9.3)(yaml@2.8.3):
     dependencies:
       bundle-require: 5.1.0(esbuild@0.27.2)
       cac: 6.7.14
@@ -5955,16 +6277,16 @@ snapshots:
       fix-dts-default-cjs-exports: 1.0.1
       joycon: 3.1.1
       picocolors: 1.1.1
-      postcss-load-config: 6.0.1(jiti@2.6.1)(postcss@8.5.6)(yaml@2.8.2)
+      postcss-load-config: 6.0.1(jiti@2.6.1)(postcss@8.5.8)(yaml@2.8.3)
       resolve-from: 5.0.0
-      rollup: 4.57.0
+      rollup: 4.60.0
       source-map: 0.7.6
       sucrase: 3.35.1
       tinyexec: 0.3.2
       tinyglobby: 0.2.15
       tree-kill: 1.2.2
     optionalDependencies:
-      postcss: 8.5.6
+      postcss: 8.5.8
       typescript: 5.9.3
     transitivePeerDependencies:
       - jiti
@@ -6014,7 +6336,7 @@ snapshots:
 
   undici-types@7.16.0: {}
 
-  undici@7.19.1: {}
+  undici@7.24.5: {}
 
   unicode-emoji-modifier-base@1.0.0: {}
 
@@ -6047,24 +6369,27 @@ snapshots:
 
   vary@1.1.2: {}
 
-  vite@7.3.1(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2):
+  vite@8.0.5(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3):
     dependencies:
-      esbuild: 0.27.2
-      fdir: 6.5.0(picomatch@4.0.3)
-      picomatch: 4.0.3
-      postcss: 8.5.6
-      rollup: 4.57.0
+      lightningcss: 1.32.0
+      picomatch: 4.0.4
+      postcss: 8.5.8
+      rolldown: 1.0.0-rc.12(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)
       tinyglobby: 0.2.15
     optionalDependencies:
       '@types/node': 25.0.10
+      esbuild: 0.27.2
       fsevents: 2.3.3
       jiti: 2.6.1
-      yaml: 2.8.2
+      yaml: 2.8.3
+    transitivePeerDependencies:
+      - '@emnapi/core'
+      - '@emnapi/runtime'
 
-  vitest@4.0.18(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2):
+  vitest@4.0.18(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3):
     dependencies:
       '@vitest/expect': 4.0.18
-      '@vitest/mocker': 4.0.18(vite@7.3.1(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2))
+      '@vitest/mocker': 4.0.18(vite@8.0.5(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3))
       '@vitest/pretty-format': 4.0.18
       '@vitest/runner': 4.0.18
       '@vitest/snapshot': 4.0.18
@@ -6075,20 +6400,23 @@ snapshots:
       magic-string: 0.30.21
       obug: 2.1.1
       pathe: 2.0.3
-      picomatch: 4.0.3
+      picomatch: 4.0.4
       std-env: 3.10.0
       tinybench: 2.9.0
       tinyexec: 1.0.2
       tinyglobby: 0.2.15
       tinyrainbow: 3.0.3
-      vite: 7.3.1(@types/node@25.0.10)(jiti@2.6.1)(yaml@2.8.2)
+      vite: 8.0.5(@emnapi/core@1.9.2)(@emnapi/runtime@1.9.2)(@types/node@25.0.10)(esbuild@0.27.2)(jiti@2.6.1)(yaml@2.8.3)
       why-is-node-running: 2.3.0
     optionalDependencies:
       '@types/node': 25.0.10
     transitivePeerDependencies:
+      - '@emnapi/core'
+      - '@emnapi/runtime'
+      - '@vitejs/devtools'
+      - esbuild
       - jiti
       - less
-      - lightningcss
       - msw
       - sass
       - sass-embedded
@@ -6135,7 +6463,7 @@ snapshots:
 
   y18n@5.0.8: {}
 
-  yaml@2.8.2: {}
+  yaml@2.8.3: {}
 
   yargs-parser@20.2.9: {}
 
diff --git a/src/benchmark/gold-standard.test.ts b/src/benchmark/gold-standard.test.ts
new file mode 100644
index 0000000..4b66cb9
--- /dev/null
+++ b/src/benchmark/gold-standard.test.ts
@@ -0,0 +1,139 @@
+import { describe, expect, it } from 'vitest';
+
+import {
+  calculateCorrelation,
+  EXCELLENT_PROMPTS,
+  FAIR_PROMPTS,
+  getExpectedScoreRange,
+  getScoreTier,
+  GOLD_STANDARD_PROMPTS,
+  GOOD_PROMPTS,
+  POOR_PROMPTS,
+  scoresMatch,
+} from './gold-standard.js';
+
+describe('gold-standard', () => {
+  describe('getScoreTier', () => {
+    it('returns excellent for scores >= 85', () => {
+      expect(getScoreTier(85)).toBe('excellent');
+      expect(getScoreTier(100)).toBe('excellent');
+    });
+
+    it('returns good for scores 70-84', () => {
+      expect(getScoreTier(70)).toBe('good');
+      expect(getScoreTier(84)).toBe('good');
+    });
+
+    it('returns fair for scores 50-69', () => {
+      expect(getScoreTier(50)).toBe('fair');
+      expect(getScoreTier(69)).toBe('fair');
+    });
+
+    it('returns poor for scores < 50', () => {
+      expect(getScoreTier(49)).toBe('poor');
+      expect(getScoreTier(0)).toBe('poor');
+    });
+  });
+
+  describe('scoresMatch', () => {
+    it('returns true for scores in same tier', () => {
+      expect(scoresMatch(85, 95)).toBe(true);
+      expect(scoresMatch(70, 80)).toBe(true);
+      expect(scoresMatch(50, 60)).toBe(true);
+      expect(scoresMatch(10, 40)).toBe(true);
+    });
+
+    it('returns false for scores in different tiers', () => {
+      expect(scoresMatch(85, 70)).toBe(false);
+      expect(scoresMatch(50, 49)).toBe(false);
+    });
+  });
+
+  describe('calculateCorrelation', () => {
+    it('returns 1 for perfect positive correlation', () => {
+      expect(calculateCorrelation([1, 2, 3], [1, 2, 3])).toBeCloseTo(1);
+    });
+
+    it('returns -1 for perfect negative correlation', () => {
+      expect(calculateCorrelation([1, 2, 3], [3, 2, 1])).toBeCloseTo(-1);
+    });
+
+    it('returns 0 for no correlation', () => {
+      expect(calculateCorrelation([1, 1, 1], [1, 2, 3])).toBeCloseTo(0);
+    });
+
+    it('returns 0 for empty arrays', () => {
+      expect(calculateCorrelation([], [])).toBe(0);
+    });
+
+    it('returns 0 for mismatched lengths', () => {
+      expect(calculateCorrelation([1, 2], [1, 2, 3])).toBe(0);
+    });
+  });
+
+  describe('getExpectedScoreRange', () => {
+    it('returns 80-100 for 0 issues', () => {
+      expect(getExpectedScoreRange(0)).toEqual({ min: 80, max: 100 });
+    });
+
+    it('returns 60-85 for 1 issue', () => {
+      expect(getExpectedScoreRange(1)).toEqual({ min: 60, max: 85 });
+    });
+
+    it('returns 40-70 for 2 issues', () => {
+      expect(getExpectedScoreRange(2)).toEqual({ min: 40, max: 70 });
+    });
+
+    it('returns 0-55 for 3+ issues', () => {
+      expect(getExpectedScoreRange(3)).toEqual({ min: 0, max: 55 });
+      expect(getExpectedScoreRange(5)).toEqual({ min: 0, max: 55 });
+    });
+  });
+
+  describe('GOLD_STANDARD_PROMPTS', () => {
+    it('has 50 total prompts', () => {
+      expect(GOLD_STANDARD_PROMPTS).toHaveLength(50);
+    });
+
+    it('has 10 excellent prompts', () => {
+      expect(EXCELLENT_PROMPTS).toHaveLength(10);
+    });
+
+    it('has 15 good prompts', () => {
+      expect(GOOD_PROMPTS).toHaveLength(15);
+    });
+
+    it('has 15 fair prompts', () => {
+      expect(FAIR_PROMPTS).toHaveLength(15);
+    });
+
+    it('has 10 poor prompts', () => {
+      expect(POOR_PROMPTS).toHaveLength(10);
+    });
+
+    it('excellent prompts have scores >= 85', () => {
+      for (const prompt of EXCELLENT_PROMPTS) {
+        expect(prompt.expectedScore).toBeGreaterThanOrEqual(85);
+      }
+    });
+
+    it('good prompts have scores 70-84', () => {
+      for (const prompt of GOOD_PROMPTS) {
+        expect(prompt.expectedScore).toBeGreaterThanOrEqual(70);
+        expect(prompt.expectedScore).toBeLessThan(85);
+      }
+    });
+
+    it('fair prompts have scores 38-69', () => {
+      for (const prompt of FAIR_PROMPTS) {
+        expect(prompt.expectedScore).toBeLessThan(70);
+      }
+    });
+
+    it('poor prompts have scores < 25', () => {
+      for (const prompt of POOR_PROMPTS) {
+        expect(prompt.expectedScore).toBeLessThan(25);
+      }
+    });
+  });
+});
diff --git a/src/benchmark/gold-standard.ts b/src/benchmark/gold-standard.ts
new file mode 100644
index 0000000..4e6b7a5
--- /dev/null
+++ b/src/benchmark/gold-standard.ts
@@ -0,0 +1,375 @@
+/**
+ * Gold Standard Benchmark Dataset
+ */
+
+export type GoldStandardPrompt = {
+  readonly text: string;
+  readonly expectedScore: number;
+  readonly expectedIssues: readonly string[];
+  readonly rationale: string;
+};
+
+export type ScoreTier = 'excellent' | 'good' | 'fair' | 'poor';
+
+export function getScoreTier(score: number): ScoreTier {
+  if (score >= 85) return 'excellent';
+  if (score >= 70) return 'good';
+  if (score >= 50) return 'fair';
+  return 'poor';
+}
+
+export function scoresMatch(actual: number, expected: number): boolean {
+  return getScoreTier(actual) === getScoreTier(expected);
+}
+
+export function calculateCorrelation(
+  predictions: readonly number[],
+  expectations: readonly number[],
+): number {
+  if (predictions.length !== expectations.length || predictions.length === 0)
+    return 0;
+  const n = predictions.length;
+  const meanPred = predictions.reduce((a, b) => a + b, 0) / n;
+  const meanExp = expectations.reduce((a, b) => a + b, 0) / n;
+  let num = 0,
+    denomPred = 0,
+    denomExp = 0;
+  for (let i = 0; i < n; i++) {
+    const dp = (predictions[i] ?? 0) - meanPred,
+      de = (expectations[i] ?? 0) - meanExp;
+    num += dp * de;
+    denomPred += dp * dp;
+    denomExp += de * de;
+  }
+  const denom = Math.sqrt(denomPred * denomExp);
+  return denom === 0 ? 0 : num / denom;
+}
+
+export const EXCELLENT_PROMPTS: readonly GoldStandardPrompt[] = [
+  {
+    text: 'Fix the null pointer exception in auth.ts line 45 where user.email is undefined when called from the password reset flow',
+    expectedScore: 95,
+    expectedIssues: [],
+    rationale: 'Specific file, line, error, and context',
+  },
+  {
+    text: 'Add error handling to validateUser() in src/utils/auth.ts for database timeout after 5s',
+    expectedScore: 92,
+    expectedIssues: [],
+    rationale: 'Clear goal, function, path, constraint',
+  },
+  {
+    text: 'Refactor calculateTotal() to use reduce instead of forEach, maintaining return type number',
+    expectedScore: 88,
+    expectedIssues: [],
+    rationale: 'Specific function, clear transformation',
+  },
+  {
+    text: 'Write unit tests for UserService.createUser covering: success, duplicate email, invalid password',
+    expectedScore: 90,
+    expectedIssues: [],
+    rationale: 'Clear method, specific test cases',
+  },
+  {
+    text: 'Update docker-compose.yml to add Redis container for session caching on port 6379',
+    expectedScore: 91,
+    expectedIssues: [],
+    rationale: 'Specific file, clear addition',
+  },
+  {
+    text: 'Debug why /api/users returns 500 when email contains + sign - check URL encoding',
+    expectedScore: 89,
+    expectedIssues: [],
+    rationale: 'Specific endpoint, exact condition',
+  },
+  {
+    text: 'Implement rate limiting for login: max 5 attempts per IP per minute, return 429',
+    expectedScore: 93,
+    expectedIssues: [],
+    rationale: 'Exact limits specified',
+  },
+  {
+    text: 'Add TypeScript types in src/types/api.ts matching docs/openapi.yaml',
+    expectedScore: 87,
+    expectedIssues: [],
+    rationale: 'Clear file paths, reference doc',
+  },
+  {
+    text: 'Optimize getUserOrders() to use single JOIN instead of N+1 - currently 3s for 100 orders',
+    expectedScore: 90,
+    expectedIssues: [],
+    rationale: 'Specific function, perf baseline',
+  },
+  {
+    text: 'Fix race condition in useAuth hook where logout completes before token refresh resolves',
+    expectedScore: 88,
+    expectedIssues: [],
+    rationale: 'Specific hook, exact race described',
+  },
+];
+
+export const GOOD_PROMPTS: readonly GoldStandardPrompt[] = [
+  {
+    text: 'Add validation to signup form for email format and password length',
+    expectedScore: 78,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing file path and rules',
+  },
+  {
+    text: 'Fix the bug where users are logged out after page refresh',
+    expectedScore: 72,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'No file paths or errors',
+  },
+  {
+    text: 'Improve performance of product listing - too slow to load',
+    expectedScore: 70,
+    expectedIssues: ['missing-technical-details', 'insufficient-constraints'],
+    rationale: 'Vague "too slow"',
+  },
+  {
+    text: 'Add loading spinner while API call is in progress in dashboard',
+    expectedScore: 75,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing file path',
+  },
+  {
+    text: 'Update button styles to match new design system colors',
+    expectedScore: 73,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing colors and files',
+  },
+  {
+    text: 'Fix TypeScript errors in auth module after v5 upgrade',
+    expectedScore: 74,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing specific errors',
+  },
+  {
+    text: 'Add error messages to form when validation fails',
+    expectedScore: 71,
+    expectedIssues: ['missing-technical-details', 'no-context'],
+    rationale: 'Which form?',
+  },
+  {
+    text: 'Implement dark mode toggle in settings page',
+    expectedScore: 76,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing implementation details',
+  },
+  {
+    text: 'Write documentation for API endpoints in readme',
+    expectedScore: 72,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing specifics',
+  },
+  {
+    text: 'Add caching to expensive database queries',
+    expectedScore: 74,
+    expectedIssues: ['missing-technical-details', 'insufficient-constraints'],
+    rationale: 'Which queries?',
+  },
+  {
+    text: 'Create reusable Modal component for confirmation dialogs',
+    expectedScore: 77,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing props interface',
+  },
+  {
+    text: 'Fix mobile layout issues on checkout page',
+    expectedScore: 73,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing breakpoints',
+  },
+  {
+    text: 'Add pagination to users list in admin panel',
+    expectedScore: 78,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing page size',
+  },
+  {
+    text: 'Implement search for blog posts',
+    expectedScore: 75,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing criteria',
+  },
+  {
+    text: 'Add input sanitization to prevent XSS in comments',
+    expectedScore: 79,
+    expectedIssues: ['missing-technical-details'],
+    rationale: 'Missing files',
+  },
+];
+
+export const FAIR_PROMPTS: readonly GoldStandardPrompt[] = [
+  {
+    text: 'Fix the bug in the login',
+    expectedScore: 55,
+    expectedIssues: ['no-context', 'missing-technical-details'],
+    rationale: 'What bug?',
+  },
+  {
+    text: 'Make the page faster',
+    expectedScore: 50,
+    expectedIssues: ['vague', 'no-context', 'missing-technical-details'],
+    rationale: 'Which page?',
+  },
+  {
+    text: 'Add some tests',
+    expectedScore: 45,
+    expectedIssues: ['vague', 'no-context', 'insufficient-constraints'],
+    rationale: 'What tests?',
+  },
+  {
+    text: 'Clean up the code',
+    expectedScore: 42,
+    expectedIssues: ['vague', 'no-context'],
+    rationale: 'What code?',
+  },
+  {
+    text: 'There is an error somewhere',
+    expectedScore: 38,
+    expectedIssues: ['vague', 'no-context', 'missing-technical-details'],
+    rationale: 'No location',
+  },
+  {
+    text: 'Improve this component',
+    expectedScore: 48,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Which component?',
+  },
+  {
+    text: 'Update the dependencies',
+    expectedScore: 55,
+    expectedIssues: ['no-goal', 'insufficient-constraints'],
+    rationale: 'Which ones?',
+  },
+  {
+    text: 'Something is wrong with authentication',
+    expectedScore: 40,
+    expectedIssues: ['vague', 'no-context', 'missing-technical-details'],
+    rationale: 'No details',
+  },
+  {
+    text: 'Handle edge cases',
+    expectedScore: 44,
+    expectedIssues: ['vague', 'no-context'],
+    rationale: 'What cases?',
+  },
+  {
+    text: 'Make it more secure',
+    expectedScore: 46,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'What concerns?',
+  },
+  {
+    text: 'Add error handling',
+    expectedScore: 52,
+    expectedIssues: ['no-context', 'missing-technical-details'],
+    rationale: 'Where?',
+  },
+  {
+    text: 'Refactor this file',
+    expectedScore: 45,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Which file?',
+  },
+  {
+    text: 'The button does not work',
+    expectedScore: 50,
+    expectedIssues: ['no-context', 'missing-technical-details'],
+    rationale: 'Which button?',
+  },
+  {
+    text: 'Add logging',
+    expectedScore: 54,
+    expectedIssues: ['no-context', 'insufficient-constraints'],
+    rationale: 'Where?',
+  },
+  {
+    text: 'Check the API',
+    expectedScore: 48,
+    expectedIssues: ['vague', 'no-goal', 'no-context'],
+    rationale: 'Check what?',
+  },
+];
+
+export const POOR_PROMPTS: readonly GoldStandardPrompt[] = [
+  {
+    text: 'Fix it',
+    expectedScore: 15,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'No information',
+  },
+  {
+    text: 'Help',
+    expectedScore: 10,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Single word',
+  },
+  {
+    text: 'Make it work',
+    expectedScore: 20,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'What?',
+  },
+  {
+    text: 'Debug',
+    expectedScore: 12,
+    expectedIssues: ['vague', 'no-context', 'no-goal', 'imperative'],
+    rationale: 'Just a verb',
+  },
+  {
+    text: '???',
+    expectedScore: 5,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Not a request',
+  },
+  {
+    text: 'Code',
+    expectedScore: 8,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Single word',
+  },
+  {
+    text: 'Do the thing',
+    expectedScore: 18,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'What thing?',
+  },
+  {
+    text: 'Error',
+    expectedScore: 10,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Just a word',
+  },
+  {
+    text: 'Please',
+    expectedScore: 5,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Says nothing',
+  },
+  {
+    text: 'This',
+    expectedScore: 5,
+    expectedIssues: ['vague', 'no-context', 'no-goal'],
+    rationale: 'Meaningless',
+  },
+];
+
+export const GOLD_STANDARD_PROMPTS: readonly GoldStandardPrompt[] = [
+  ...EXCELLENT_PROMPTS,
+  ...GOOD_PROMPTS,
+  ...FAIR_PROMPTS,
+  ...POOR_PROMPTS,
+];
+
+export function getExpectedScoreRange(issueCount: number): {
+  min: number;
+  max: number;
+} {
+  if (issueCount === 0) return { min: 80, max: 100 };
+  if (issueCount === 1) return { min: 60, max: 85 };
+  if (issueCount === 2) return { min: 40, max: 70 };
+  return { min: 0, max: 55 };
+}
diff --git a/src/benchmark/index.ts b/src/benchmark/index.ts
new file mode 100644
index 0000000..536a767
--- /dev/null
+++ b/src/benchmark/index.ts
@@ -0,0 +1,16 @@
+/**
+ * Benchmark module exports.
+ */
+export {
+  calculateCorrelation,
+  EXCELLENT_PROMPTS,
+  FAIR_PROMPTS,
+  getExpectedScoreRange,
+  getScoreTier,
+  GOLD_STANDARD_PROMPTS,
+  type GoldStandardPrompt,
+  GOOD_PROMPTS,
+  POOR_PROMPTS,
+  scoresMatch,
+  type ScoreTier,
+} from './gold-standard.js';
diff --git a/src/cli.ts b/src/cli.ts
index cbdb6f3..4425fe8 100644
--- a/src/cli.ts
+++ b/src/cli.ts
@@ -1734,7 +1734,8 @@ export async function cli(): Promise<void> {
       const results: AnalysisResult[] = [];
 
       for (const group of groups) {
-        const prompts = group.prompts.map((p) => p.content);
+        const filteredPrompts = group.prompts.filter((p) => !p.isConfirmation);
+        const prompts = filteredPrompts.map((p) => p.content);
         const result = await analyzeWithProgress(
           provider,
           prompts,
@@ -1743,7 +1744,7 @@ export async function cli(): Promise<void> {
           config.rules,
           isJsonMode,
           args.noCache,
-          group.prompts,
+          filteredPrompts,
         );
         results.push(result);
       }
@@ -1824,7 +1825,19 @@ export async function cli(): Promise<void> {
       }
     } else {
       // Single-day analysis
-      const prompts = logResult.prompts.map((p) => p.content);
+      const filteredPrompts = logResult.prompts.filter(
+        (p) => !p.isConfirmation,
+      );
+      const confirmationsExcluded =
+        logResult.prompts.length - filteredPrompts.length;
+      if (confirmationsExcluded > 0) {
+        console.log(
+          chalk.dim(
+            `  Excluded ${String(confirmationsExcluded)} confirmation(s) (short replies to assistant questions)`,
+          ),
+        );
+      }
+      const prompts = filteredPrompts.map((p) => p.content);
       const date = args.date;
       let result: AnalysisResult | EnhancedAnalysisResult =
         await analyzeWithProgress(
@@ -1835,7 +1848,7 @@ export async function cli(): Promise<void> {
           config.rules,
           isJsonMode,
           args.noCache,
-          logResult.prompts,
+          filteredPrompts,
         );
 
       // Enrich with enhanced analytics if requested
diff --git a/src/core/analyzer.test.ts b/src/core/analyzer.test.ts
index 941faad..a7bed18 100644
--- a/src/core/analyzer.test.ts
+++ b/src/core/analyzer.test.ts
@@ -461,7 +461,7 @@ describe('mergeBatchResults', () => {
     });
 
     const pattern = merged.patterns.find((p) => p.id === 'p1');
-    expect(pattern?.frequency).toBe(5); // (4 + 6) / 2 = 5
+    expect(pattern?.frequency).toBe(10); // 4 + 6 = 10 (sum across batches)
   });
 
   it('should take max severity for duplicate patterns', () => {
diff --git a/src/core/analyzer.ts b/src/core/analyzer.ts
index 912c6a4..dbd480d 100644
--- a/src/core/analyzer.ts
+++ b/src/core/analyzer.ts
@@ -572,7 +572,7 @@ export function mergeBatchResults(
       const severities = group.map((p) => p.severity);
       const allExamples = group.flatMap((p) => Array.from(p.examples));
 
-      const mergedFrequency = Math.round(average(frequencies));
+      const mergedFrequency = sum(frequencies);
       const mergedSeverity = maxSeverity(severities);
       const mergedExamples = limitExamples(
         allExamples,
diff --git a/src/core/log-reader.test.ts b/src/core/log-reader.test.ts
index 8240cc2..dbbc6f1 100644
--- a/src/core/log-reader.test.ts
+++ b/src/core/log-reader.test.ts
@@ -15,7 +15,9 @@ import {
   getProjects,
   groupByDay,
   isClaudeMessage,
+  isConfirmationMessage,
   parseDate,
+  parseLogEntry,
   readLogs,
 } from './log-reader.js';
 
@@ -148,13 +150,15 @@ describe('readLogs', () => {
     const result = await readLogs();
 
     expect(result.prompts).toHaveLength(1);
-    expect(result.prompts[0]).toEqual({
-      content: 'Test prompt',
-      timestamp: '2025-01-23T14:30:00.000Z',
-      sessionId: 'abc123',
-      project: 'my-project',
-      date: '2025-01-23',
-    });
+    expect(result.prompts[0]).toEqual(
+      expect.objectContaining({
+        content: 'Test prompt',
+        timestamp: '2025-01-23T14:30:00.000Z',
+        sessionId: 'abc123',
+        project: 'my-project',
+        date: '2025-01-23',
+      }),
+    );
   });
 
   it('handles malformed JSON lines gracefully', async () => {
@@ -907,3 +911,249 @@ describe('readLogs with filters', () => {
     expect(firstWarning).toContain('Invalid date');
   });
 });
+
+describe('parseLogEntry', () => {
+  it('should parse a user message with uuid and parentUuid', () => {
+    const line = JSON.stringify({
+      uuid: 'user-1',
+      parentUuid: 'system-1',
+      type: 'user',
+      message: { role: 'user', content: 'si' },
+      timestamp: '2026-04-07T10:00:00.000Z',
+      sessionId: 'sess-1',
+      cwd: '/test',
+    });
+
+    const result = parseLogEntry(line);
+    expect(result).not.toBeNull();
+    expect(result?.uuid).toBe('user-1');
+    expect(result?.entry.type).toBe('user');
+    expect(result?.entry.parentUuid).toBe('system-1');
+    expect(result?.entry.contentTail).toBe('si');
+  });
+
+  it('should extract text from assistant array content', () => {
+    const line = JSON.stringify({
+      uuid: 'asst-1',
+      type: 'assistant',
+      message: {
+        role: 'assistant',
+        content: [{ type: 'text', text: 'Should I proceed with the changes?' }],
+      },
+      timestamp: '2026-04-07T10:00:00.000Z',
+    });
+
+    const result = parseLogEntry(line);
+    expect(result?.entry.contentTail).toBe(
+      'Should I proceed with the changes?',
+    );
+  });
+
+  it('should return null for lines without uuid', () => {
+    const line = JSON.stringify({
+      type: 'system',
+      message: { content: 'test' },
+    });
+    expect(parseLogEntry(line)).toBeNull();
+  });
+
+  it('should return null for invalid JSON', () => {
+    expect(parseLogEntry('not json')).toBeNull();
+  });
+
+  it('should return null for empty lines', () => {
+    expect(parseLogEntry('')).toBeNull();
+    expect(parseLogEntry('  ')).toBeNull();
+  });
+});
+
+describe('isConfirmationMessage', () => {
+  function buildIndex(
+    entries: {
+      uuid: string;
+      type: string;
+      contentTail: string | null;
+      parentUuid: string | null;
+    }[],
+  ): Map<
+    string,
+    { type: string; contentTail: string | null; parentUuid: string | null }
+  > {
+    const index = new Map<
+      string,
+      { type: string; contentTail: string | null; parentUuid: string | null }
+    >();
+    for (const e of entries) {
+      index.set(e.uuid, {
+        type: e.type,
+        contentTail: e.contentTail,
+        parentUuid: e.parentUuid,
+      });
+    }
+    return index;
+  }
+
+  it('should detect confirmation via assistant→system→user chain', () => {
+    const index = buildIndex([
+      {
+        uuid: 'asst-1',
+        type: 'assistant',
+        contentTail: 'Should I proceed?',
+        parentUuid: null,
+      },
+      {
+        uuid: 'sys-1',
+        type: 'system',
+        contentTail: null,
+        parentUuid: 'asst-1',
+      },
+    ]);
+
+    expect(isConfirmationMessage('si', 'sys-1', index)).toBe(true);
+  });
+
+  it('should return false when assistant does not end with question mark', () => {
+    const index = buildIndex([
+      {
+        uuid: 'asst-1',
+        type: 'assistant',
+        contentTail: 'Done. Changes applied.',
+        parentUuid: null,
+      },
+      {
+        uuid: 'sys-1',
+        type: 'system',
+        contentTail: null,
+        parentUuid: 'asst-1',
+      },
+    ]);
+
+    expect(isConfirmationMessage('ok', 'sys-1', index)).toBe(false);
+  });
+
+  it('should return false for long messages even with question parent', () => {
+    const index = buildIndex([
+      {
+        uuid: 'asst-1',
+        type: 'assistant',
+        contentTail: 'Should I proceed?',
+        parentUuid: null,
+      },
+    ]);
+
+    expect(
+      isConfirmationMessage(
+        'si, hazme un refactor completo de auth',
+        'asst-1',
+        index,
+      ),
+    ).toBe(false);
+  });
+
+  it('should return false when there is no parentUuid', () => {
+    const index = buildIndex([]);
+    expect(isConfirmationMessage('si', null, index)).toBe(false);
+  });
+
+  it('should return false when chain exceeds max hops', () => {
+    const index = buildIndex([
+      { uuid: 's1', type: 'system', contentTail: null, parentUuid: 's2' },
+      { uuid: 's2', type: 'system', contentTail: null, parentUuid: 's3' },
+      { uuid: 's3', type: 'system', contentTail: null, parentUuid: 's4' },
+      { uuid: 's4', type: 'system', contentTail: null, parentUuid: 's5' },
+      { uuid: 's5', type: 'system', contentTail: null, parentUuid: 's6' },
+      {
+        uuid: 's6',
+        type: 'assistant',
+        contentTail: 'Question?',
+        parentUuid: null,
+      },
+    ]);
+
+    expect(isConfirmationMessage('si', 's1', index)).toBe(false);
+  });
+
+  it('should detect direct assistant→user confirmation', () => {
+    const index = buildIndex([
+      {
+        uuid: 'asst-1',
+        type: 'assistant',
+        contentTail: '¿Procedo con los cambios?',
+        parentUuid: null,
+      },
+    ]);
+
+    expect(isConfirmationMessage('si', 'asst-1', index)).toBe(true);
+  });
+
+  it('should return false when parentUuid is not in index', () => {
+    const index = buildIndex([]);
+    expect(isConfirmationMessage('si', 'unknown-uuid', index)).toBe(false);
+  });
+});
+
+describe('readJsonlFile confirmation integration', () => {
+  it('should mark short replies to assistant questions as confirmations', async () => {
+    mockExistsSync.mockReturnValue(true);
+    mockGlob.mockResolvedValue(['/mock/.claude/projects/my-project/log.jsonl']);
+
+    const lines = [
+      JSON.stringify({
+        uuid: 'asst-1',
+        parentUuid: null,
+        type: 'assistant',
+        message: {
+          role: 'assistant',
+          content: [
+            { type: 'text', text: '¿Quieres que proceda con los cambios?' },
+          ],
+        },
+        timestamp: '2026-04-07T10:00:00.000Z',
+        sessionId: 'sess-1',
+        cwd: '/test',
+      }),
+      JSON.stringify({
+        uuid: 'sys-1',
+        parentUuid: 'asst-1',
+        type: 'system',
+        message: { role: 'system', content: 'stop_hook_summary' },
+        timestamp: '2026-04-07T10:00:01.000Z',
+        sessionId: 'sess-1',
+        cwd: '/test',
+      }),
+      JSON.stringify({
+        uuid: 'user-1',
+        parentUuid: 'sys-1',
+        type: 'user',
+        message: { role: 'user', content: 'si' },
+        timestamp: '2026-04-07T10:00:02.000Z',
+        sessionId: 'sess-1',
+        cwd: '/test',
+      }),
+      JSON.stringify({
+        uuid: 'user-2',
+        parentUuid: 'user-1',
+        type: 'user',
+        message: {
+          role: 'user',
+          content: 'Refactoriza el módulo de autenticación',
+        },
+        timestamp: '2026-04-07T10:01:00.000Z',
+        sessionId: 'sess-1',
+        cwd: '/test',
+      }),
+    ].join('\n');
+
+    mockReadFile.mockResolvedValue(lines);
+
+    const result = await readLogs();
+
+    expect(result.prompts).toHaveLength(2);
+    expect(result.prompts[0]?.content).toBe('si');
+    expect(result.prompts[0]?.isConfirmation).toBe(true);
+    expect(result.prompts[1]?.content).toBe(
+      'Refactoriza el módulo de autenticación',
+    );
+    expect(result.prompts[1]?.isConfirmation).toBe(false);
+  });
+});
diff --git a/src/core/log-reader.ts b/src/core/log-reader.ts
index 274cf4e..e4f399b 100644
--- a/src/core/log-reader.ts
+++ b/src/core/log-reader.ts
@@ -217,6 +217,95 @@ export function extractContent(message: ClaudeMessage): string {
   return message.message.content;
 }
 
+// =============================================================================
+// Confirmation Detection
+// =============================================================================
+
+const MAX_CONFIRMATION_LENGTH = 20;
+const MAX_CHAIN_HOPS = 5;
+
+type IndexEntry = {
+  readonly type: string;
+  readonly contentTail: string | null;
+  readonly parentUuid: string | null;
+};
+
+/**
+ * Lightweight parser that extracts minimal fields from any log entry type.
+ * Used to build the message index for confirmation detection.
+ * Separate from parseLine/isClaudeMessage to handle assistant array content.
+ */
+export function parseLogEntry(
+  line: string,
+): { uuid: string; entry: IndexEntry } | null {
+  if (!line.trim()) return null;
+
+  try {
+    const obj = JSON.parse(line) as Record<string, unknown>;
+    const uuid = obj['uuid'];
+    if (typeof uuid !== 'string') return null;
+
+    const type = typeof obj['type'] === 'string' ? obj['type'] : 'unknown';
+    const parentUuid =
+      typeof obj['parentUuid'] === 'string' ? obj['parentUuid'] : null;
+
+    let contentTail: string | null = null;
+    const message = obj['message'] as Record<string, unknown> | undefined;
+    if (message) {
+      const content = message['content'];
+      if (typeof content === 'string') {
+        contentTail = content.slice(-100);
+      } else if (Array.isArray(content)) {
+        const textBlock = content.find(
+          (block: unknown) =>
+            typeof block === 'object' &&
+            block !== null &&
+            (block as Record<string, unknown>)['type'] === 'text',
+        ) as Record<string, unknown> | undefined;
+        const text = textBlock?.['text'];
+        if (typeof text === 'string') {
+          contentTail = text.slice(-100);
+        }
+      }
+    }
+
+    return { uuid, entry: { type, contentTail, parentUuid } };
+  } catch {
+    return null;
+  }
+}
+
+/**
+ * Detects if a short user message is a confirmation response to an assistant question.
+ * Walks the parentUuid chain backwards to find the nearest assistant message
+ * and checks if it ends with a question mark.
+ */
+export function isConfirmationMessage(
+  content: string,
+  parentUuid: string | null,
+  index: ReadonlyMap<string, IndexEntry>,
+): boolean {
+  if (content.trim().length > MAX_CONFIRMATION_LENGTH) return false;
+
+  let currentUuid = parentUuid;
+  for (let hop = 0; hop < MAX_CHAIN_HOPS; hop++) {
+    if (!currentUuid) return false;
+    const entry = index.get(currentUuid);
+    if (!entry) return false;
+
+    if (entry.type === 'assistant') {
+      return entry.contentTail?.trimEnd().endsWith('?') === true;
+    }
+    currentUuid = entry.parentUuid;
+  }
+
+  return false;
+}
+
+// =============================================================================
+// File Reading
+// =============================================================================
+
 /**
  * Reads and parses a single JSONL file.
  *
@@ -236,6 +325,16 @@ async function readJsonlFile(filePath: string): Promise<{
     const lines = content.split('\n');
     let skippedLines = 0;
 
+    // First pass: build message index for confirmation detection
+    const messageIndex = new Map<string, IndexEntry>();
+    for (const line of lines) {
+      const parsed = parseLogEntry(line);
+      if (parsed) {
+        messageIndex.set(parsed.uuid, parsed.entry);
+      }
+    }
+
+    // Second pass: extract user prompts with confirmation detection
     for (let i = 0; i < lines.length; i++) {
       const line = lines[i] ?? '';
       const lineNumber = i + 1;
@@ -257,12 +356,21 @@ async function readJsonlFile(filePath: string): Promise<{
         continue;
       }
 
+      // Get parentUuid for this user message from the index
+      const logEntry = parseLogEntry(line);
+      const userParentUuid = logEntry?.entry.parentUuid ?? null;
+
       prompts.push({
         content: extractedContent,
         timestamp: message.timestamp,
         sessionId: message.sessionId,
         project: projectName,
         date: extractDate(message.timestamp),
+        isConfirmation: isConfirmationMessage(
+          extractedContent,
+          userParentUuid,
+          messageIndex,
+        ),
       });
     }
 
diff --git a/src/core/model-suggester.test.ts b/src/core/model-suggester.test.ts
index ef9aaf9..337d15b 100644
--- a/src/core/model-suggester.test.ts
+++ b/src/core/model-suggester.test.ts
@@ -11,7 +11,7 @@ describe('model-suggester', () => {
     it('should suggest qwen2.5:14b when available with 8GB+ RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'qwen2.5:14b', size: 9000000000 },
         { name: 'mistral:7b', size: 4000000000 },
       ];
@@ -25,40 +25,40 @@ describe('model-suggester', () => {
     it('should suggest mistral:7b when qwen not available but mistral is, with 8GB+ RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'mistral:7b', size: 4000000000 },
       ];
 
       const result = suggestBestModel(systemInfo, models);
 
       expect(result.suggestedModel).toBe('mistral:7b');
-      expect(result.reason).toContain('Small Schema');
+      expect(result.reason).toContain('Full Schema');
     });
 
-    it('should suggest gemma3:4b when available with low RAM', () => {
+    it('should suggest gemma4:e4b when available with low RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 4 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'mistral:7b', size: 4000000000 },
       ];
 
       const result = suggestBestModel(systemInfo, models);
 
-      expect(result.suggestedModel).toBe('gemma3:4b');
-      expect(result.reason).toContain('lightweight');
+      expect(result.suggestedModel).toBe('gemma4:e4b');
+      expect(result.reason).toContain('128K context');
     });
 
-    it('should fallback to gemma3:4b when no models available', () => {
+    it('should fallback to gemma4:e4b when no models available', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [];
 
       const result = suggestBestModel(systemInfo, models);
 
-      expect(result.suggestedModel).toBe('gemma3:4b');
+      expect(result.suggestedModel).toBe('gemma4:e4b');
       expect(result.reason).toContain('default');
     });
 
-    it('should fallback to gemma3:4b when only unknown models available', () => {
+    it('should fallback to gemma4:e4b when only unknown models available', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
         { name: 'unknown-model:latest', size: 5000000000 },
@@ -66,13 +66,13 @@ describe('model-suggester', () => {
 
       const result = suggestBestModel(systemInfo, models);
 
-      expect(result.suggestedModel).toBe('gemma3:4b');
+      expect(result.suggestedModel).toBe('gemma4:e4b');
     });
 
     it('should prefer qwen2.5:14b over mistral:7b when both available with high RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 32 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'mistral:7b', size: 4000000000 },
         { name: 'qwen2.5:14b', size: 9000000000 },
       ];
@@ -85,7 +85,7 @@ describe('model-suggester', () => {
     it('should handle models with tags (e.g., :latest)', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'mistral:7b-instruct', size: 4000000000 },
       ];
 
@@ -94,21 +94,21 @@ describe('model-suggester', () => {
       expect(result.suggestedModel).toContain('mistral');
     });
 
-    it('should suggest gemma3:4b with exactly 8GB RAM when only gemma3:4b available', () => {
+    it('should suggest gemma4:e4b with exactly 8GB RAM when only gemma4:e4b available', () => {
       const systemInfo: SystemInfo = { ramGB: 8 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
       ];
 
       const result = suggestBestModel(systemInfo, models);
 
-      expect(result.suggestedModel).toBe('gemma3:4b');
+      expect(result.suggestedModel).toBe('gemma4:e4b');
     });
 
     it('should suggest mistral:7b with exactly 8GB RAM when available', () => {
       const systemInfo: SystemInfo = { ramGB: 8 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'mistral:7b', size: 4000000000 },
       ];
 
@@ -122,7 +122,7 @@ describe('model-suggester', () => {
     it('should suggest mistral:7b with 8GB+ RAM when not installed', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
       ];
 
       const result = getInstallationSuggestion(systemInfo, models);
@@ -132,21 +132,21 @@ describe('model-suggester', () => {
       expect(result?.benefits).toContain('Better analysis quality');
     });
 
-    it('should suggest gemma3:4b with 4-7GB RAM when not installed', () => {
+    it('should suggest gemma4:e4b with 4-7GB RAM when not installed', () => {
       const systemInfo: SystemInfo = { ramGB: 6 };
       const models: OllamaModelInfo[] = [];
 
       const result = getInstallationSuggestion(systemInfo, models);
 
       expect(result).not.toBeNull();
-      expect(result?.suggestedModel).toBe('gemma3:4b');
-      expect(result?.benefits).toContain('Fast and lightweight');
+      expect(result?.suggestedModel).toBe('gemma4:e4b');
+      expect(result?.benefits).toContain('128K context');
     });
 
     it('should not suggest when mistral:7b already installed with 8GB+ RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'mistral:7b', size: 4000000000 },
       ];
 
@@ -158,7 +158,7 @@ describe('model-suggester', () => {
     it('should not suggest when qwen2.5:14b already installed with 8GB+ RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 16 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
         { name: 'qwen2.5:14b', size: 9000000000 },
       ];
 
@@ -167,10 +167,10 @@ describe('model-suggester', () => {
       expect(result).toBeNull();
     });
 
-    it('should not suggest when gemma3:4b already installed with 4-7GB RAM', () => {
+    it('should not suggest when gemma4:e4b already installed with 4-7GB RAM', () => {
       const systemInfo: SystemInfo = { ramGB: 6 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
       ];
 
       const result = getInstallationSuggestion(systemInfo, models);
@@ -201,7 +201,7 @@ describe('model-suggester', () => {
     it('should suggest with exactly 8GB RAM when mistral not installed', () => {
       const systemInfo: SystemInfo = { ramGB: 8 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
       ];
 
       const result = getInstallationSuggestion(systemInfo, models);
@@ -210,14 +210,14 @@ describe('model-suggester', () => {
       expect(result?.suggestedModel).toBe('mistral:7b');
     });
 
-    it('should suggest with exactly 4GB RAM when gemma3:4b not installed', () => {
+    it('should suggest with exactly 4GB RAM when gemma4:e4b not installed', () => {
       const systemInfo: SystemInfo = { ramGB: 4 };
       const models: OllamaModelInfo[] = [];
 
       const result = getInstallationSuggestion(systemInfo, models);
 
       expect(result).not.toBeNull();
-      expect(result?.suggestedModel).toBe('gemma3:4b');
+      expect(result?.suggestedModel).toBe('gemma4:e4b');
     });
   });
 
@@ -227,7 +227,7 @@ describe('model-suggester', () => {
       const models: OllamaModelInfo[] = [];
 
       const modelSuggestion = suggestBestModel(systemInfo, models);
-      expect(modelSuggestion.suggestedModel).toBe('gemma3:4b');
+      expect(modelSuggestion.suggestedModel).toBe('gemma4:e4b');
 
       const installSuggestion = getInstallationSuggestion(systemInfo, models);
       expect(installSuggestion).not.toBeNull();
@@ -247,11 +247,11 @@ describe('model-suggester', () => {
     it('should handle very low RAM values', () => {
       const systemInfo: SystemInfo = { ramGB: 1 };
       const models: OllamaModelInfo[] = [
-        { name: 'gemma3:4b:latest', size: 2000000000 },
+        { name: 'gemma4:e4b:latest', size: 5000000000 },
       ];
 
       const modelSuggestion = suggestBestModel(systemInfo, models);
-      expect(modelSuggestion.suggestedModel).toBe('gemma3:4b');
+      expect(modelSuggestion.suggestedModel).toBe('gemma4:e4b');
 
       const installSuggestion = getInstallationSuggestion(systemInfo, models);
       expect(installSuggestion).toBeNull(); // Too little RAM
diff --git a/src/core/model-suggester.ts b/src/core/model-suggester.ts
index 792b2fc..76ee055 100644
--- a/src/core/model-suggester.ts
+++ b/src/core/model-suggester.ts
@@ -38,9 +38,9 @@ const LOW_RAM_THRESHOLD = 4;
  * Priority order (with 8GB+ RAM):
  * 1. qwen2.5:14b if available
  * 2. mistral:7b if available
- * 3. llama3.2 (default fallback)
+ * 3. gemma4:e4b (default fallback)
  *
- * With less than 8GB RAM, always suggests llama3.2 for reliability.
+ * With less than 8GB RAM, always suggests gemma4:e4b for reliability.
  *
  * @param systemInfo - System hardware information
  * @param models - List of available Ollama models
@@ -58,22 +58,22 @@ export function suggestBestModel(
   systemInfo: SystemInfo,
   models: OllamaModelInfo[],
 ): ModelRecommendation {
-  // With low RAM, always suggest llama3.2 for reliability
+  // With low RAM, suggest gemma4:e4b for reliability
   if (systemInfo.ramGB < HIGH_RAM_THRESHOLD) {
-    if (isModelAvailable('gemma3:4b', models)) {
+    if (isModelAvailable('gemma4:e4b', models)) {
       return {
-        suggestedModel: 'gemma3:4b',
-        reason: 'Fast and lightweight model, optimal for your system',
+        suggestedModel: 'gemma4:e4b',
+        reason: 'Fast model with 128K context and native function calling',
       };
     }
     return {
-      suggestedModel: 'gemma3:4b',
+      suggestedModel: 'gemma4:e4b',
       reason: 'Recommended default model (needs installation)',
     };
   }
 
   // With 8GB+ RAM, suggest better models if available
-  // Priority: qwen2.5:14b > mistral:7b > llama3.2
+  // Priority: qwen2.5:14b > mistral:7b > gemma4:e4b
 
   if (isModelAvailable('qwen2.5:14b', models)) {
     return {
@@ -85,20 +85,20 @@ export function suggestBestModel(
   if (isModelAvailable('mistral:7b', models)) {
     return {
       suggestedModel: 'mistral:7b',
-      reason: 'Better analysis quality with Small Schema',
+      reason: 'Good analysis quality with Full Schema',
     };
   }
 
-  // Fallback to llama3.2
-  if (isModelAvailable('gemma3:4b', models)) {
+  // Fallback to gemma4:e4b
+  if (isModelAvailable('gemma4:e4b', models)) {
     return {
-      suggestedModel: 'gemma3:4b',
-      reason: 'Fast and reliable default model',
+      suggestedModel: 'gemma4:e4b',
+      reason: 'Fast and reliable default model with 128K context',
     };
   }
 
   return {
-    suggestedModel: 'gemma3:4b',
+    suggestedModel: 'gemma4:e4b',
     reason: 'Recommended default model (needs installation)',
   };
 }
@@ -107,7 +107,7 @@ export function suggestBestModel(
  * Determines if we should suggest installing a better model.
  * Logic:
  * - 8GB+ RAM: suggest mistral:7b if not installed (and qwen2.5:14b not installed)
- * - 4-7GB RAM: suggest llama3.2 if not installed
+ * - 4-7GB RAM: suggest gemma4:e4b if not installed
  * - < 4GB RAM: no suggestion (insufficient resources)
  *
  * @param systemInfo - System hardware information
@@ -139,7 +139,7 @@ export function getInstallationSuggestion(
     if (!hasMistral && !hasQwen) {
       return {
         suggestedModel: 'mistral:7b',
-        benefits: 'Better analysis quality with Small Schema',
+        benefits: 'Better analysis quality with Full Schema',
         installCommand: 'ollama pull mistral:7b',
       };
     }
@@ -147,14 +147,14 @@ export function getInstallationSuggestion(
     return null; // Already has optimal model
   }
 
-  // Low RAM (4-7GB): suggest llama3.2 if not installed
-  const hasLlama = isModelAvailable('gemma3:4b', models);
+  // Low RAM (4-7GB): suggest gemma4:e4b if not installed
+  const hasDefault = isModelAvailable('gemma4:e4b', models);
 
-  if (!hasLlama) {
+  if (!hasDefault) {
     return {
-      suggestedModel: 'gemma3:4b',
-      benefits: 'Fast and lightweight model for your system',
-      installCommand: 'ollama pull llama3.2',
+      suggestedModel: 'gemma4:e4b',
+      benefits: '128K context and native function calling',
+      installCommand: 'ollama pull gemma4:e4b',
     };
   }
 
diff --git a/src/core/reminder.test.ts b/src/core/reminder.test.ts
index 2ae1adf..da99f96 100644
--- a/src/core/reminder.test.ts
+++ b/src/core/reminder.test.ts
@@ -31,7 +31,7 @@ vi.mock('../utils/env.js', () => ({
   getEnvConfig: vi.fn(() => ({
     reminder: '7d',
     services: ['ollama'],
-    ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+    ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
     anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
     google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
   })),
@@ -264,7 +264,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: 'never',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
@@ -279,7 +279,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: '7d',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
@@ -299,7 +299,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: '7d',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
@@ -319,7 +319,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: '7d',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
@@ -339,7 +339,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: '14d',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
@@ -359,7 +359,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: '30d',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
@@ -379,7 +379,7 @@ describe('shouldShowReminder', () => {
     vi.mocked(getEnvConfig).mockReturnValue({
       reminder: 'unknown',
       services: ['ollama'],
-      ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+      ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
       anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
       google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
     });
diff --git a/src/core/results-storage.test.ts b/src/core/results-storage.test.ts
index b0e038d..b216eb4 100644
--- a/src/core/results-storage.test.ts
+++ b/src/core/results-storage.test.ts
@@ -73,7 +73,7 @@ const mockMetadata: Omit<PromptResultMetadata, 'promptHash' | 'analyzedAt'> = {
   date: '2025-12-26',
   project: 'test-project',
   provider: 'ollama',
-  model: 'gemma3:4b',
+  model: 'gemma4:e4b',
   schemaType: 'full',
 };
 
@@ -104,14 +104,14 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
     const hash2 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
@@ -123,14 +123,14 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('prompt 1', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
     const hash2 = getPromptResultHash('prompt 2', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
@@ -141,14 +141,14 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
     const hash2 = getPromptResultHash('test prompt', {
       date: '2025-12-27',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
@@ -159,14 +159,14 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
     const hash2 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project2',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
@@ -177,7 +177,7 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
@@ -195,14 +195,14 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
     const hash2 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: 'project1',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'minimal',
     });
 
@@ -213,14 +213,14 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: undefined,
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
     const hash2 = getPromptResultHash('test prompt', {
       date: '2025-12-26',
       project: undefined,
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
@@ -235,7 +235,7 @@ describe('getPromptResultHash', () => {
     const hash1 = getPromptResultHash('test', {
       date: '2025-12-26',
       project: 'test',
-      model: 'gemma3:4b',
+      model: 'gemma4:e4b',
       schemaType: 'full',
     });
 
diff --git a/src/core/semantic-validator.test.ts b/src/core/semantic-validator.test.ts
new file mode 100644
index 0000000..ea6a7a0
--- /dev/null
+++ b/src/core/semantic-validator.test.ts
@@ -0,0 +1,274 @@
+import { describe, expect, it } from 'vitest';
+
+import type { AnalysisResult } from '../types/index.js';
+import {
+  autoCorrectResult,
+  validateSemantics,
+  type ValidationResult,
+} from './semantic-validator.js';
+
+// Helper to create a valid base result
+function createResult(overrides: Partial<AnalysisResult> = {}): AnalysisResult {
+  return {
+    date: '2025-01-15',
+    patterns: [],
+    stats: { totalPrompts: 10, promptsWithIssues: 0, overallScore: 90 },
+    topSuggestion: 'Be more specific',
+    ...overrides,
+  };
+}
+
+// Helper to create a pattern
+function createPattern(overrides: Record<string, unknown> = {}) {
+  return {
+    id: 'vague',
+    name: 'Vague Request',
+    frequency: 3,
+    severity: 'high' as const,
+    examples: ['help me'],
+    suggestion: 'Be specific',
+    beforeAfter: { before: 'help me', after: 'Help me fix X in Y' },
+    ...overrides,
+  };
+}
+
+describe('validateSemantics', () => {
+  it('returns valid for a well-formed result', () => {
+    const result = createResult({
+      patterns: [createPattern()],
+      stats: { totalPrompts: 10, promptsWithIssues: 3, overallScore: 70 },
+    });
+    const validation = validateSemantics(result, ['help me', 'fix bug']);
+    expect(validation.valid).toBe(true);
+    expect(validation.issues).toHaveLength(0);
+  });
+
+  it('detects score-mismatch when score is outside expected range for issue count', () => {
+    const result = createResult({
+      patterns: [
+        createPattern(),
+        createPattern({ id: 'no-context' }),
+        createPattern({ id: 'too-broad' }),
+      ],
+      stats: { totalPrompts: 10, promptsWithIssues: 7, overallScore: 95 },
+    });
+    const validation = validateSemantics(result, ['help me']);
+    expect(validation.issues.some((i) => i.type === 'score-mismatch')).toBe(
+      true,
+    );
+  });
+
+  it('detects empty-patterns with low score', () => {
+    const result = createResult({
+      patterns: [],
+      stats: { totalPrompts: 10, promptsWithIssues: 0, overallScore: 50 },
+    });
+    const validation = validateSemantics(result, ['some prompt']);
+    expect(validation.issues.some((i) => i.type === 'empty-patterns')).toBe(
+      true,
+    );
+  });
+
+  it('detects duplicate-issues', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({ id: 'vague' }),
+        createPattern({ id: 'vague' }),
+      ],
+    });
+    const validation = validateSemantics(result, ['help me']);
+    expect(validation.issues.some((i) => i.type === 'duplicate-issues')).toBe(
+      true,
+    );
+  });
+
+  it('detects example-not-found when examples do not match prompts', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({
+          examples: ['this example does not exist in any prompt at all'],
+        }),
+      ],
+    });
+    const validation = validateSemantics(result, [
+      'completely different text here',
+    ]);
+    expect(validation.issues.some((i) => i.type === 'example-not-found')).toBe(
+      true,
+    );
+  });
+
+  it('detects stats-inconsistent when promptsWithIssues exceeds totalPrompts', () => {
+    const result = createResult({
+      stats: { totalPrompts: 3, promptsWithIssues: 12, overallScore: 6 },
+    });
+    const validation = validateSemantics(result, ['a', 'b', 'c']);
+    const issue = validation.issues.find(
+      (i) => i.type === 'stats-inconsistent',
+    );
+    expect(issue).toBeDefined();
+    expect(issue?.severity).toBe('error');
+    expect(validation.valid).toBe(false);
+  });
+
+  it('detects unknown-pattern-id for IDs not in taxonomy', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({ id: 'vague-project-name', name: 'Vague Project Name' }),
+      ],
+      stats: { totalPrompts: 3, promptsWithIssues: 2, overallScore: 5 },
+    });
+    const validation = validateSemantics(result, ['help me']);
+    expect(validation.issues.some((i) => i.type === 'unknown-pattern-id')).toBe(
+      true,
+    );
+  });
+
+  it('does not flag valid taxonomy IDs', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({ id: 'vague' }),
+        createPattern({ id: 'no-context', name: 'Missing Context' }),
+      ],
+      stats: { totalPrompts: 10, promptsWithIssues: 5, overallScore: 5 },
+    });
+    const validation = validateSemantics(result, ['help me']);
+    expect(
+      validation.issues.filter((i) => i.type === 'unknown-pattern-id'),
+    ).toHaveLength(0);
+  });
+
+  it('detects placeholder-example in pattern examples', () => {
+    const result = createResult({
+      patterns: [createPattern({ examples: ['example prompt 1'] })],
+      stats: { totalPrompts: 3, promptsWithIssues: 2, overallScore: 5 },
+    });
+    const validation = validateSemantics(result, ['example prompt 1']);
+    expect(
+      validation.issues.some((i) => i.type === 'placeholder-example'),
+    ).toBe(true);
+  });
+});
+
+describe('autoCorrectResult', () => {
+  it('adjusts score to expected range midpoint', () => {
+    const result = createResult({
+      patterns: [
+        createPattern(),
+        createPattern({ id: 'no-context' }),
+        createPattern({ id: 'too-broad' }),
+      ],
+      stats: { totalPrompts: 10, promptsWithIssues: 7, overallScore: 95 },
+    });
+    const validation = validateSemantics(result, ['help me']);
+    const corrected = autoCorrectResult(result, validation);
+    expect(corrected.stats.overallScore).toBeLessThan(95);
+  });
+
+  it('removes duplicate patterns', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({ id: 'vague' }),
+        createPattern({ id: 'vague' }),
+      ],
+    });
+    const validation: ValidationResult = {
+      valid: true,
+      issues: [
+        {
+          type: 'duplicate-issues',
+          message: 'Duplicate: vague',
+          severity: 'warning',
+        },
+      ],
+    };
+    const corrected = autoCorrectResult(result, validation);
+    expect(corrected.patterns).toHaveLength(1);
+  });
+
+  it('clamps promptsWithIssues to totalPrompts', () => {
+    const result = createResult({
+      stats: { totalPrompts: 3, promptsWithIssues: 12, overallScore: 6 },
+    });
+    const validation: ValidationResult = {
+      valid: false,
+      issues: [
+        {
+          type: 'stats-inconsistent',
+          message: 'promptsWithIssues exceeds totalPrompts',
+          severity: 'error',
+        },
+      ],
+    };
+    const corrected = autoCorrectResult(result, validation);
+    expect(corrected.stats.promptsWithIssues).toBe(3);
+    expect(corrected.stats.totalPrompts).toBe(3);
+  });
+
+  it('maps unknown pattern IDs to taxonomy', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({
+          id: 'vague-project-name',
+          name: 'Vague Project Name',
+          severity: 'high',
+          suggestion: 'custom suggestion',
+        }),
+      ],
+    });
+    const validation: ValidationResult = {
+      valid: true,
+      issues: [
+        {
+          type: 'unknown-pattern-id',
+          message: 'Unknown: vague-project-name',
+          severity: 'warning',
+        },
+      ],
+    };
+    const corrected = autoCorrectResult(result, validation);
+    expect(corrected.patterns[0]?.id).toBe('vague');
+    expect(corrected.patterns[0]?.name).toBe('Vague Request');
+  });
+
+  it('deduplicates patterns after remapping IDs', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({ id: 'vague-project-name', name: 'Vague Project' }),
+        createPattern({ id: 'vague', name: 'Vague Request' }),
+      ],
+    });
+    const validation: ValidationResult = {
+      valid: true,
+      issues: [
+        {
+          type: 'unknown-pattern-id',
+          message: 'Unknown: vague-project-name',
+          severity: 'warning',
+        },
+      ],
+    };
+    const corrected = autoCorrectResult(result, validation);
+    expect(corrected.patterns.filter((p) => p.id === 'vague')).toHaveLength(1);
+  });
+
+  it('removes placeholder examples', () => {
+    const result = createResult({
+      patterns: [
+        createPattern({ examples: ['example prompt 1', 'real user prompt'] }),
+      ],
+    });
+    const validation: ValidationResult = {
+      valid: true,
+      issues: [
+        {
+          type: 'placeholder-example',
+          message: 'Placeholder detected',
+          severity: 'warning',
+        },
+      ],
+    };
+    const corrected = autoCorrectResult(result, validation);
+    expect(corrected.patterns[0]?.examples).toEqual(['real user prompt']);
+  });
+});
diff --git a/src/core/semantic-validator.ts b/src/core/semantic-validator.ts
new file mode 100644
index 0000000..eeff3a3
--- /dev/null
+++ b/src/core/semantic-validator.ts
@@ -0,0 +1,308 @@
+/**
+ * Semantic Validation for Analysis Results
+ *
+ * Post-processing validation to catch logical inconsistencies
+ * that JSON schema validation cannot detect.
+ */
+
+import { getExpectedScoreRange } from '../benchmark/gold-standard.js';
+import { ISSUE_TAXONOMY } from '../providers/schemas.js';
+import type { AnalysisResult } from '../types/index.js';
+import { logger } from '../utils/logger-base.js';
+
+export type ValidationIssue = {
+  readonly type:
+    | 'score-mismatch'
+    | 'example-not-found'
+    | 'empty-patterns'
+    | 'duplicate-issues'
+    | 'stats-inconsistent'
+    | 'unknown-pattern-id'
+    | 'placeholder-example';
+  readonly message: string;
+  readonly severity: 'warning' | 'error';
+};
+
+export type ValidationResult = {
+  readonly valid: boolean;
+  readonly issues: readonly ValidationIssue[];
+  readonly adjustedResult?: AnalysisResult;
+};
+
+/**
+ * Validates an analysis result for semantic consistency.
+ * Checks that scores correlate with issue counts and examples exist in prompts.
+ */
+export function validateSemantics(
+  result: AnalysisResult,
+  originalPrompts: readonly string[],
+): ValidationResult {
+  const issues: ValidationIssue[] = [];
+
+  // 1. Check score-to-issues correlation
+  const { patterns } = result;
+  const issueCount = patterns.length;
+  const expectedRange = getExpectedScoreRange(issueCount);
+  const actualScore = result.stats.overallScore;
+
+  if (actualScore < expectedRange.min || actualScore > expectedRange.max) {
+    issues.push({
+      type: 'score-mismatch',
+      message: `Score ${String(actualScore)} is inconsistent with ${String(issueCount)} issues (expected ${String(expectedRange.min)}-${String(expectedRange.max)})`,
+      severity: 'warning',
+    });
+  }
+
+  // 2. Check for empty patterns with low score
+  if (issueCount === 0 && actualScore < 70) {
+    issues.push({
+      type: 'empty-patterns',
+      message: `No patterns detected but score is ${String(actualScore)} (expected 70+)`,
+      severity: 'warning',
+    });
+  }
+
+  // 3. Check for duplicate issue IDs
+  if (patterns.length > 0) {
+    const seenIds = new Set<string>();
+    for (const pattern of patterns) {
+      if (seenIds.has(pattern.id)) {
+        issues.push({
+          type: 'duplicate-issues',
+          message: `Duplicate pattern ID: ${pattern.id}`,
+          severity: 'warning',
+        });
+      }
+      seenIds.add(pattern.id);
+    }
+  }
+
+  // 4. Verify examples are substrings of original prompts
+  for (const pattern of patterns) {
+    const { examples } = pattern;
+    for (const example of examples) {
+      const found = originalPrompts.some(
+        (prompt) =>
+          prompt.includes(example) ||
+          example.includes(prompt.slice(0, Math.min(50, prompt.length))),
+      );
+      if (!found && example.length > 10) {
+        issues.push({
+          type: 'example-not-found',
+          message: `Example "${example.slice(0, 30)}..." not found in prompts`,
+          severity: 'warning',
+        });
+      }
+    }
+  }
+
+  // 5. Check stats consistency: promptsWithIssues must not exceed totalPrompts
+  if (result.stats.promptsWithIssues > result.stats.totalPrompts) {
+    issues.push({
+      type: 'stats-inconsistent',
+      message: `promptsWithIssues (${String(result.stats.promptsWithIssues)}) exceeds totalPrompts (${String(result.stats.totalPrompts)})`,
+      severity: 'error',
+    });
+  }
+
+  // 6. Check for unknown pattern IDs not in taxonomy
+  const validIds = Object.keys(ISSUE_TAXONOMY);
+  for (const pattern of result.patterns) {
+    if (!validIds.includes(pattern.id)) {
+      issues.push({
+        type: 'unknown-pattern-id',
+        message: `Unknown pattern ID: "${pattern.id}" (valid: ${validIds.join(', ')})`,
+        severity: 'warning',
+      });
+    }
+  }
+
+  // 7. Check for placeholder examples in patterns
+  const PLACEHOLDER_TEXTS = [
+    'example prompt 1',
+    'example prompt 2',
+    'example prompt 3',
+    'original prompt',
+    'improved prompt',
+  ];
+  for (const pattern of result.patterns) {
+    for (const example of pattern.examples) {
+      if (PLACEHOLDER_TEXTS.some((p) => example.toLowerCase().includes(p))) {
+        issues.push({
+          type: 'placeholder-example',
+          message: `Placeholder example detected: "${example.slice(0, 40)}"`,
+          severity: 'warning',
+        });
+      }
+    }
+  }
+
+  // Log warnings for debugging
+  if (issues.length > 0) {
+    logger.debug(
+      `Semantic validation found ${String(issues.length)} issue(s)`,
+      'validator',
+    );
+    for (const issue of issues) {
+      logger.debug(`  [${issue.type}] ${issue.message}`, 'validator');
+    }
+  }
+
+  return {
+    valid: issues.filter((i) => i.severity === 'error').length === 0,
+    issues,
+  };
+}
+
+/**
+ * Finds the closest taxonomy ID for an unknown pattern ID.
+ * Uses substring and keyword matching as heuristics.
+ */
+function findClosestTaxonomyId(
+  unknownId: string,
+  validIds: readonly string[],
+): string | undefined {
+  const normalized = unknownId.toLowerCase();
+
+  // Direct substring match: check if any valid ID is contained in the unknown ID
+  for (const validId of validIds) {
+    if (normalized.includes(validId.replace('-', ''))) return validId;
+    if (normalized.includes(validId)) return validId;
+  }
+
+  // Keyword-based matching
+  const keywordMap: Record<string, string> = {
+    vague: 'vague',
+    context: 'no-context',
+    broad: 'too-broad',
+    goal: 'no-goal',
+    imperative: 'imperative',
+    command: 'imperative',
+    technical: 'missing-technical-details',
+    detail: 'missing-technical-details',
+    priorit: 'unclear-priorities',
+    constraint: 'insufficient-constraints',
+    specif: 'vague',
+    unclear: 'vague',
+    missing: 'no-context',
+  };
+
+  for (const [keyword, taxonomyId] of Object.entries(keywordMap)) {
+    if (normalized.includes(keyword)) return taxonomyId;
+  }
+
+  return undefined;
+}
+
+/**
+ * Attempts to fix common semantic issues in results.
+ * Returns adjusted result if fixes were applied.
+ */
+export function autoCorrectResult(
+  result: AnalysisResult,
+  validation: ValidationResult,
+): AnalysisResult {
+  let adjusted = result;
+
+  for (const issue of validation.issues) {
+    if (issue.type === 'score-mismatch') {
+      // Adjust score to match issue count
+      const { patterns } = result;
+      const issueCount = patterns.length;
+      const expectedRange = getExpectedScoreRange(issueCount);
+      const midpoint = Math.round((expectedRange.min + expectedRange.max) / 2);
+
+      if (
+        result.stats.overallScore < expectedRange.min ||
+        result.stats.overallScore > expectedRange.max
+      ) {
+        adjusted = {
+          ...adjusted,
+          stats: {
+            ...adjusted.stats,
+            overallScore: midpoint,
+          },
+        };
+        logger.debug(
+          `Auto-corrected score from ${String(result.stats.overallScore)} to ${String(midpoint)}`,
+          'validator',
+        );
+      }
+    }
+
+    if (issue.type === 'duplicate-issues') {
+      // Remove duplicate patterns
+      const seen = new Set<string>();
+      const uniquePatterns = adjusted.patterns.filter((p) => {
+        if (seen.has(p.id)) return false;
+        seen.add(p.id);
+        return true;
+      });
+      adjusted = { ...adjusted, patterns: uniquePatterns };
+    }
+
+    if (issue.type === 'stats-inconsistent') {
+      adjusted = {
+        ...adjusted,
+        stats: {
+          ...adjusted.stats,
+          promptsWithIssues: Math.min(
+            adjusted.stats.promptsWithIssues,
+            adjusted.stats.totalPrompts,
+          ),
+        },
+      };
+    }
+
+    if (issue.type === 'unknown-pattern-id') {
+      const validIds = Object.keys(ISSUE_TAXONOMY);
+      const mappedPatterns = adjusted.patterns.map((pattern) => {
+        if (validIds.includes(pattern.id)) return pattern;
+        const matchedId = findClosestTaxonomyId(pattern.id, validIds);
+        const metadata = matchedId ? ISSUE_TAXONOMY[matchedId] : undefined;
+        if (matchedId && metadata) {
+          return {
+            ...pattern,
+            id: matchedId,
+            name: metadata.name,
+            severity: metadata.severity,
+            suggestion: metadata.suggestion,
+          };
+        }
+        return pattern;
+      });
+      // Deduplicate after remapping (multiple IDs may map to same taxonomy ID)
+      const seen = new Set<string>();
+      adjusted = {
+        ...adjusted,
+        patterns: mappedPatterns.filter((p) => {
+          if (seen.has(p.id)) return false;
+          seen.add(p.id);
+          return true;
+        }),
+      };
+    }
+
+    if (issue.type === 'placeholder-example') {
+      const PLACEHOLDER_TEXTS = [
+        'example prompt 1',
+        'example prompt 2',
+        'example prompt 3',
+        'original prompt',
+        'improved prompt',
+      ];
+      adjusted = {
+        ...adjusted,
+        patterns: adjusted.patterns.map((pattern) => ({
+          ...pattern,
+          examples: pattern.examples.filter(
+            (ex) =>
+              !PLACEHOLDER_TEXTS.some((p) => ex.toLowerCase().includes(p)),
+          ),
+        })),
+      };
+    }
+  }
+
+  return adjusted;
+}
diff --git a/src/core/setup.test.ts b/src/core/setup.test.ts
index 1480b4d..292d754 100644
--- a/src/core/setup.test.ts
+++ b/src/core/setup.test.ts
@@ -68,7 +68,7 @@ describe('setup', () => {
     it('should complete successfully with ollama provider selected', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] }) // Provider selection
-        .mockResolvedValueOnce({ model: 'gemma3:4b' }) // Ollama model
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' }) // Ollama model
         .mockResolvedValueOnce({ host: 'http://localhost:11434' }) // Ollama host
         .mockResolvedValueOnce({ reminder: '7d' }) // Reminder
         .mockResolvedValueOnce({ saveToShell: true }); // Save to shell
@@ -83,10 +83,10 @@ describe('setup', () => {
       const config = await runSetup();
 
       expect(config.services).toEqual(['ollama']);
-      expect(config.ollama.model).toBe('gemma3:4b');
+      expect(config.ollama.model).toBe('gemma4:e4b');
       expect(config.ollama.host).toBe('http://localhost:11434');
       expect(process.env['HYNTX_SERVICES']).toBe('ollama');
-      expect(process.env['HYNTX_OLLAMA_MODEL']).toBe('gemma3:4b');
+      expect(process.env['HYNTX_OLLAMA_MODEL']).toBe('gemma4:e4b');
     });
 
     it('should complete successfully with anthropic provider selected', async () => {
@@ -140,7 +140,7 @@ describe('setup', () => {
     it('should handle multiple providers selected', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama', 'anthropic'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' }) // Ollama model
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' }) // Ollama model
         .mockResolvedValueOnce({ host: 'http://localhost:11434' }) // Ollama host
         .mockResolvedValueOnce({ model: 'claude-3-5-haiku-latest' }) // Anthropic model
         .mockResolvedValueOnce({ apiKey: 'sk-test-key' }) // Anthropic key
@@ -150,7 +150,7 @@ describe('setup', () => {
       const config = await runSetup();
 
       expect(config.services).toEqual(['ollama', 'anthropic']);
-      expect(config.ollama.model).toBe('gemma3:4b');
+      expect(config.ollama.model).toBe('gemma4:e4b');
       expect(config.anthropic.apiKey).toBe('sk-test-key');
       expect(process.env['HYNTX_SERVICES']).toBe('ollama,anthropic');
     });
@@ -191,7 +191,7 @@ describe('setup', () => {
       const { EXIT_CODES } = await import('../types/index.js');
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '7d' })
         .mockResolvedValueOnce({}); // User cancelled - empty response
@@ -211,7 +211,7 @@ describe('setup', () => {
     it('should set environment variables for selected providers', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '7d' })
         .mockResolvedValueOnce({ saveToShell: false });
@@ -220,14 +220,14 @@ describe('setup', () => {
 
       expect(process.env['HYNTX_SERVICES']).toBe('ollama');
       expect(process.env['HYNTX_REMINDER']).toBe('7d');
-      expect(process.env['HYNTX_OLLAMA_MODEL']).toBe('gemma3:4b');
+      expect(process.env['HYNTX_OLLAMA_MODEL']).toBe('gemma4:e4b');
       expect(process.env['HYNTX_OLLAMA_HOST']).toBe('http://localhost:11434');
     });
 
     it('should not set anthropic env vars if provider not selected', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: 'never' })
         .mockResolvedValueOnce({ saveToShell: false });
@@ -241,7 +241,7 @@ describe('setup', () => {
     it('should call saveConfigToShell when user confirms', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '7d' })
         .mockResolvedValueOnce({ saveToShell: true });
@@ -265,7 +265,7 @@ describe('setup', () => {
     it('should show manual instructions when saveConfigToShell fails', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '7d' })
         .mockResolvedValueOnce({ saveToShell: true });
@@ -285,7 +285,7 @@ describe('setup', () => {
     it('should show manual instructions when user declines save', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '7d' })
         .mockResolvedValueOnce({ saveToShell: false });
@@ -299,7 +299,7 @@ describe('setup', () => {
     it('should handle undefined saveToShell response (defaults to false)', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '7d' })
         .mockResolvedValueOnce({ saveToShell: undefined });
@@ -321,14 +321,14 @@ describe('setup', () => {
       const config = await runSetup();
 
       // Should use ENV_DEFAULTS values
-      expect(config.ollama.model).toBe('gemma3:4b');
+      expect(config.ollama.model).toBe('gemma4:e4b');
       expect(config.ollama.host).toBe('http://localhost:11434');
     });
 
     it('should handle reminder never option', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: 'never' })
         .mockResolvedValueOnce({ saveToShell: false });
@@ -343,7 +343,7 @@ describe('setup', () => {
     it('should update config when reminder is not never', async () => {
       vi.mocked(prompts)
         .mockResolvedValueOnce({ providers: ['ollama'] })
-        .mockResolvedValueOnce({ model: 'gemma3:4b' })
+        .mockResolvedValueOnce({ model: 'gemma4:e4b' })
         .mockResolvedValueOnce({ host: 'http://localhost:11434' })
         .mockResolvedValueOnce({ reminder: '30d' })
         .mockResolvedValueOnce({ saveToShell: false });
@@ -386,7 +386,7 @@ describe('setup', () => {
       const config: EnvConfig = {
         services: ['ollama'],
         reminder: '7d',
-        ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+        ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
         anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
         google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
       };
@@ -405,7 +405,7 @@ describe('setup', () => {
       const config: EnvConfig = {
         services: ['ollama'],
         reminder: 'never',
-        ollama: { model: 'gemma3:4b', host: 'http://localhost:11434' },
+        ollama: { model: 'gemma4:e4b', host: 'http://localhost:11434' },
         anthropic: { model: 'claude-3-5-haiku-latest', apiKey: '' },
         google: { model: 'gemini-2.0-flash-exp', apiKey: '' },
       };
diff --git a/src/providers/base.ts b/src/providers/base.ts
index 7d53f5f..0e66c44 100644
--- a/src/providers/base.ts
+++ b/src/providers/base.ts
@@ -29,6 +29,7 @@ const PLACEHOLDER_PATTERNS = [
   'Improved version addressing the issue',
   '<PLACEHOLDER_EXAMPLE>',
   'Human-Readable Issue Name',
+  'example prompt 1',
 ] as const;
 
 /**
diff --git a/src/providers/ollama.test.ts b/src/providers/ollama.test.ts
index a854245..13bfb69 100644
--- a/src/providers/ollama.test.ts
+++ b/src/providers/ollama.test.ts
@@ -15,10 +15,9 @@ import { detectBatchStrategy, OllamaProvider } from './ollama.js';
 
 describe('OllamaProvider', () => {
   const mockConfig: OllamaConfig = {
-    model: 'llama3.2',
+    model: 'llama3:70b',
     host: 'http://localhost:11434',
-    // Force full schema for tests to match expected response format
-    schemaOverride: 'batch',
+    // Use standard-strategy model so auto-detection selects 'full' schema
   };
 
   let provider: OllamaProvider;
@@ -54,7 +53,7 @@ describe('OllamaProvider', () => {
       global.fetch = vi.fn().mockResolvedValue({
         ok: true,
         json: async () => ({
-          models: [{ name: 'llama3.2' }, { name: 'other-model' }],
+          models: [{ name: 'llama3:70b' }, { name: 'other-model' }],
         }),
       });
 
@@ -72,7 +71,7 @@ describe('OllamaProvider', () => {
       global.fetch = vi.fn().mockResolvedValue({
         ok: true,
         json: async () => ({
-          models: [{ name: 'llama3.2:latest' }, { name: 'other-model' }],
+          models: [{ name: 'llama3:70b-instruct' }, { name: 'other-model' }],
         }),
       });
 
@@ -238,7 +237,7 @@ describe('OllamaProvider', () => {
       if (!callArgs) throw new Error('Expected fetch to be called');
       const body = JSON.parse(callArgs[1].body);
 
-      expect(body.model).toBe('llama3.2');
+      expect(body.model).toBe('llama3:70b');
       expect(body.stream).toBe(false);
       expect(body.options.temperature).toBe(0.3);
       expect(body.prompt).toContain('Test prompt');
@@ -275,7 +274,7 @@ describe('OllamaProvider', () => {
       });
 
       await expect(provider.analyze(['Test'], '2025-01-15')).rejects.toThrow(
-        'Ollama API request failed: 500 Internal Server Error',
+        'Ollama API failed: 500 Internal Server Error',
       );
     });
 
@@ -375,7 +374,7 @@ describe('OllamaProvider', () => {
         'Failed to parse response as JSON',
       );
 
-      expect(global.fetch).toHaveBeenCalledTimes(1); // No retries for parse errors
+      expect(global.fetch).toHaveBeenCalledTimes(3); // Tries all 3 temperature levels before failing
     });
 
     it('should not retry on schema validation errors', async () => {
@@ -390,7 +389,7 @@ describe('OllamaProvider', () => {
         'Response does not match expected schema',
       );
 
-      expect(global.fetch).toHaveBeenCalledTimes(1); // No retries for schema errors
+      expect(global.fetch).toHaveBeenCalledTimes(3); // Tries all 3 temperature levels before failing
     });
 
     it('should use exponential backoff for retries', async () => {
@@ -597,33 +596,44 @@ describe('OllamaProvider', () => {
   });
 
   describe('getBatchLimits', () => {
-    it('should return micro strategy limits for llama3.2', () => {
+    it('should return micro strategy limits for llama3.2 (individual mode)', () => {
       const provider = new OllamaProvider({
         model: 'llama3.2',
         host: 'http://localhost:11434',
-        // Without override, llama3.2 uses individual schema (maxPromptsPerBatch = 1)
-        // With batch override, it uses the strategy's actual limits
-        schemaOverride: 'batch',
       });
 
       const limits = provider.getBatchLimits();
 
       expect(limits.maxTokensPerBatch).toBe(500);
-      expect(limits.maxPromptsPerBatch).toBe(3);
+      // Micro models use individual schema → 1 prompt per batch
+      expect(limits.maxPromptsPerBatch).toBe(1);
       expect(limits.prioritization).toBe('longest-first');
     });
 
-    it('should return small strategy limits for mistral:7b', () => {
+    it('should return small strategy limits for mistral:7b (full schema mode)', () => {
       const provider = new OllamaProvider({
         model: 'mistral:7b',
         host: 'http://localhost:11434',
-        // Without override, mistral:7b uses individual schema (maxPromptsPerBatch = 1)
-        schemaOverride: 'batch',
       });
 
       const limits = provider.getBatchLimits();
 
       expect(limits.maxTokensPerBatch).toBe(1500);
+      // Small models use full schema → up to 10 prompts per batch
+      expect(limits.maxPromptsPerBatch).toBe(10);
+      expect(limits.prioritization).toBe('longest-first');
+    });
+
+    it('should return small strategy limits for gemma4:e4b (full schema mode)', () => {
+      const provider = new OllamaProvider({
+        model: 'gemma4:e4b',
+        host: 'http://localhost:11434',
+      });
+
+      const limits = provider.getBatchLimits();
+
+      expect(limits.maxTokensPerBatch).toBe(1500);
+      // gemma4:e4b maps to small strategy → full schema → up to 10 prompts per batch
       expect(limits.maxPromptsPerBatch).toBe(10);
       expect(limits.prioritization).toBe('longest-first');
     });
@@ -641,18 +651,17 @@ describe('OllamaProvider', () => {
       expect(limits.prioritization).toBe('longest-first');
     });
 
-    it('should return micro strategy limits for unknown models', () => {
+    it('should return micro strategy limits for unknown models (individual mode)', () => {
       const provider = new OllamaProvider({
         model: 'unknown-model',
         host: 'http://localhost:11434',
-        // Unknown models default to micro strategy with individual schema
-        schemaOverride: 'batch',
       });
 
       const limits = provider.getBatchLimits();
 
       expect(limits.maxTokensPerBatch).toBe(500);
-      expect(limits.maxPromptsPerBatch).toBe(3);
+      // Unknown models default to micro → individual schema → 1 prompt per batch
+      expect(limits.maxPromptsPerBatch).toBe(1);
       expect(limits.prioritization).toBe('longest-first');
     });
   });
@@ -699,6 +708,18 @@ describe('detectBatchStrategy', () => {
     it('should detect standard strategy for qwen2.5:14b', () => {
       expect(detectBatchStrategy('qwen2.5:14b')).toBe('standard');
     });
+
+    it('should detect micro strategy for gemma4:e2b', () => {
+      expect(detectBatchStrategy('gemma4:e2b')).toBe('micro');
+    });
+
+    it('should detect small strategy for gemma4:e4b', () => {
+      expect(detectBatchStrategy('gemma4:e4b')).toBe('small');
+    });
+
+    it('should detect standard strategy for gemma4:31b', () => {
+      expect(detectBatchStrategy('gemma4:31b')).toBe('standard');
+    });
   });
 
   describe('partial match', () => {
@@ -721,6 +742,10 @@ describe('detectBatchStrategy', () => {
     it('should detect standard strategy for mixtral-8x7b-instruct', () => {
       expect(detectBatchStrategy('mixtral-8x7b-instruct')).toBe('standard');
     });
+
+    it('should detect small strategy for gemma4:e4b:latest', () => {
+      expect(detectBatchStrategy('gemma4:e4b:latest')).toBe('small');
+    });
   });
 
   describe('default fallback', () => {
@@ -746,21 +771,27 @@ describe('detectBatchStrategy', () => {
       const micro = BATCH_STRATEGIES.micro;
       expect(micro.maxTokensPerBatch).toBe(500);
       expect(micro.maxPromptsPerBatch).toBe(3);
-      expect(micro.description).toBe('For models < 4GB');
+      expect(micro.description).toBe(
+        'Conservative: individual schema, 1 prompt at a time',
+      );
     });
 
     it('should have correct small strategy configuration', () => {
       const small = BATCH_STRATEGIES.small;
       expect(small.maxTokensPerBatch).toBe(1500);
       expect(small.maxPromptsPerBatch).toBe(10);
-      expect(small.description).toBe('For models 4-7GB');
+      expect(small.description).toBe(
+        'Balanced: full schema, up to 10 prompts per batch',
+      );
     });
 
     it('should have correct standard strategy configuration', () => {
       const standard = BATCH_STRATEGIES.standard;
       expect(standard.maxTokensPerBatch).toBe(3000);
       expect(standard.maxPromptsPerBatch).toBe(50);
-      expect(standard.description).toBe('For models > 7GB');
+      expect(standard.description).toBe(
+        'Maximum: full schema, up to 50 prompts per batch',
+      );
     });
   });
 });
diff --git a/src/providers/ollama.ts b/src/providers/ollama.ts
index b9cf079..f811702 100644
--- a/src/providers/ollama.ts
+++ b/src/providers/ollama.ts
@@ -1,13 +1,11 @@
 /**
- * Ollama AI provider implementation.
- *
- * This module provides integration with local Ollama instances for prompt analysis.
- * Features include:
- * - Availability checking with timeout and model verification
- * - Retry logic with exponential backoff for network errors
- * - Support for both raw JSON and markdown-wrapped responses
+ * Ollama AI provider implementation with improved retry and validation.
  */
 
+import {
+  autoCorrectResult,
+  validateSemantics,
+} from '../core/semantic-validator.js';
 import {
   type AnalysisProvider,
   type AnalysisResult,
@@ -30,145 +28,66 @@ import {
   SYSTEM_PROMPT_MINIMAL,
 } from './schemas.js';
 
-/**
- * Maximum number of retry attempts for network errors.
- */
 const MAX_RETRIES = 2;
-
-/**
- * Timeout for availability check (3 seconds).
- */
+const TEMPERATURE_LEVELS = [0.3, 0.1, 0.0] as const;
 const AVAILABILITY_TIMEOUT_MS = 3000;
-
-/**
- * Timeout for analysis request (60 seconds).
- */
 const ANALYSIS_TIMEOUT_MS = 60000;
-
-/**
- * Base delay for exponential backoff (1 second).
- */
 const BASE_RETRY_DELAY_MS = 1000;
 
-/**
- * Model-to-strategy mapping for known Ollama models.
- * Maps model names to their optimal batch strategy.
- */
 const MODEL_STRATEGY_MAP: Record<string, BatchStrategyType> = {
-  // Micro (< 4GB)
   'llama3.2': 'micro',
   'phi3:mini': 'micro',
   'gemma3:4b': 'micro',
   'gemma2:2b': 'micro',
-
-  // Small (4-7GB)
+  'gemma4:e2b': 'micro',
   'mistral:7b': 'small',
   'llama3:8b': 'small',
   'codellama:7b': 'small',
-
-  // Standard (> 7GB)
+  'gemma4:e4b': 'small',
   'llama3:70b': 'standard',
   mixtral: 'standard',
   'qwen2.5:14b': 'standard',
+  'gemma4:31b': 'standard',
 };
 
-/**
- * Detects the optimal batch strategy for a given model.
- * Uses exact and partial matching against known model names.
- *
- * @param modelName - Name of the Ollama model
- * @returns Batch strategy type
- *
- * @example
- * ```typescript
- * detectBatchStrategy('llama3.2') // 'micro'
- * detectBatchStrategy('llama3.2:latest') // 'micro' (partial match)
- * detectBatchStrategy('unknown-model') // 'micro' (safe default)
- * ```
- */
 export function detectBatchStrategy(modelName: string): BatchStrategyType {
-  // Check exact match first
-  if (MODEL_STRATEGY_MAP[modelName]) {
-    return MODEL_STRATEGY_MAP[modelName];
-  }
-
-  // Check partial match (e.g., "llama3.2:latest" matches "llama3.2")
+  if (MODEL_STRATEGY_MAP[modelName]) return MODEL_STRATEGY_MAP[modelName];
   for (const [pattern, strategy] of Object.entries(MODEL_STRATEGY_MAP)) {
-    if (modelName.includes(pattern)) {
-      return strategy;
-    }
+    if (modelName.includes(pattern)) return strategy;
   }
-
-  // Default to micro for unknown models (safest)
   return 'micro';
 }
 
-/**
- * Ollama provider for local AI analysis.
- * Implements the AnalysisProvider interface for Ollama instances.
- */
 export class OllamaProvider implements AnalysisProvider {
   public readonly name = 'Ollama';
   private readonly config: OllamaConfig;
   private readonly batchStrategy: BatchStrategyType;
   private readonly schemaType: SchemaType;
 
-  /**
-   * Creates a new OllamaProvider instance.
-   * Automatically detects the optimal batch strategy and schema type based on model name.
-   *
-   * @param config - Ollama configuration with model and host
-   */
   constructor(config: OllamaConfig) {
     this.config = config;
     this.batchStrategy = detectBatchStrategy(config.model);
     this.schemaType = this.selectSchemaType();
-
     const strategy = BATCH_STRATEGIES[this.batchStrategy];
     logger.debug(
-      `Detected batch strategy: ${this.batchStrategy} (${strategy.description}), schema type: ${this.schemaType}`,
+      `Detected strategy: ${this.batchStrategy} (${strategy.description}), schema: ${this.schemaType}`,
       'ollama',
     );
   }
 
-  /**
-   * Selects the appropriate schema type based on model size and user override.
-   * Micro and small models use individual schema (hybrid approach) by default.
-   * Standard models use full schema for detailed analysis.
-   * User can override via CLI --analysis-mode flag.
-   *
-   * @returns Schema type identifier
-   */
   private selectSchemaType(): SchemaType {
-    // Check for user override first
-    if (this.config.schemaOverride) {
-      return this.config.schemaOverride === 'individual'
-        ? 'individual'
-        : 'full';
+    // Only override auto-detection when explicitly requesting 'individual'
+    // 'batch' mode should respect the model's strategy-based schema selection
+    if (this.config.schemaOverride === 'individual') {
+      return 'individual';
     }
-
-    // Auto-select based on model size
-    // Micro and small models use batch-individual hybrid for better accuracy
-    // This combines batching performance with individual result clarity
-    return ['micro', 'small'].includes(this.batchStrategy)
-      ? 'individual'
-      : 'full';
+    return this.batchStrategy === 'micro' ? 'individual' : 'full';
   }
 
-  /**
-   * Returns dynamic batch limits based on detected model strategy.
-   * When using individual schema, processes one prompt at a time for reliability.
-   *
-   * @returns Provider limits with model-specific constraints
-   */
   public getBatchLimits(): ProviderLimits {
     const strategy = BATCH_STRATEGIES[this.batchStrategy];
-
-    // Individual schema processes one prompt at a time
-    // This is necessary because small models struggle to return arrays
     const maxPromptsPerBatch =
       this.schemaType === 'individual' ? 1 : strategy.maxPromptsPerBatch;
-
     return {
       maxTokensPerBatch: strategy.maxTokensPerBatch,
       maxPromptsPerBatch,
@@ -176,76 +95,47 @@ export class OllamaProvider implements AnalysisProvider {
     };
   }
 
-  /**
-   * Checks if the Ollama service is available and has the required model.
-   * Uses a 3-second timeout to avoid hanging on unreachable services.
-   *
-   * @returns Promise that resolves to true if available, false otherwise
-   */
   public async isAvailable(): Promise<boolean> {
     logger.debug(`Connecting to Ollama at ${this.config.host}`, 'ollama');
-
     try {
       const controller = new AbortController();
       const timeoutId = setTimeout(() => {
         controller.abort();
       }, AVAILABILITY_TIMEOUT_MS);
-
       const response = await fetch(`${this.config.host}/api/tags`, {
         signal: controller.signal,
       });
-
       clearTimeout(timeoutId);
-
       if (!response.ok) {
         logger.debug(
-          `Ollama API returned ${String(response.status)}`,
+          `Ollama API returned ${response.status.toString()}`,
           'ollama',
         );
         return false;
       }
-
       const data = (await response.json()) as { models?: { name: string }[] };
-
       if (!data.models || !Array.isArray(data.models)) {
         logger.debug('Ollama returned invalid model list', 'ollama');
         return false;
       }
-
-      // Check if the configured model is available
-      const modelAvailable = data.models.some((model) =>
-        model.name.includes(this.config.model),
+      const modelAvailable = data.models.some((m) =>
+        m.name.includes(this.config.model),
+      );
+      logger.debug(
+        modelAvailable
+          ? `Model ${this.config.model} available`
+          : `Model ${this.config.model} not found`,
+        'ollama',
       );
-
-      if (modelAvailable) {
-        logger.debug(
-          `Model ${this.config.model} available (${String(data.models.length)} models found)`,
-          'ollama',
-        );
-      } else {
-        logger.debug(
-          `Model ${this.config.model} not found in available models`,
-          'ollama',
-        );
-      }
-
       return modelAvailable;
     } catch {
-      // Network errors, timeouts, or JSON parse errors all indicate unavailability
       logger.debug('Ollama connection failed', 'ollama');
       return false;
     }
   }
 
   /**
-   * Analyzes prompts using the Ollama service.
-   * Implements retry logic with exponential backoff for network errors.
-   *
-   * @param prompts - Array of prompt strings to analyze
-   * @param date - Date context for the analysis
-   * @param context - Optional project context for analysis
-   * @returns Promise resolving to AnalysisResult
-   * @throws Error if analysis fails after retries or if response is invalid
+   * Analyzes prompts with temperature fallback and semantic validation.
    */
   public async analyze(
     prompts: readonly string[],
@@ -265,91 +155,110 @@ export class OllamaProvider implements AnalysisProvider {
           : SYSTEM_PROMPT_FULL;
 
     let lastError: Error | undefined;
+    let lastParseError: Error | undefined;
 
+    // Outer loop: network retries
     for (let attempt = 0; attempt <= MAX_RETRIES; attempt++) {
-      try {
-        const controller = new AbortController();
-        const timeoutId = setTimeout(() => {
-          controller.abort();
-        }, ANALYSIS_TIMEOUT_MS);
-
-        const response = await fetch(`${this.config.host}/api/generate`, {
-          method: 'POST',
-          headers: {
-            'Content-Type': 'application/json',
-          },
-          body: JSON.stringify({
-            model: this.config.model,
-            prompt: userPrompt,
-            system: systemPrompt,
-            stream: false,
-            format: 'json',
-            options: {
-              temperature: 0.3,
-            },
-          }),
-          signal: controller.signal,
-        });
-
-        clearTimeout(timeoutId);
-
-        if (!response.ok) {
-          const status = String(response.status);
-          throw new Error(
-            `Ollama API request failed: ${status} ${response.statusText}`,
+      // Inner loop: temperature fallback for parse errors
+      for (const temperature of TEMPERATURE_LEVELS) {
+        try {
+          const controller = new AbortController();
+          const timeoutId = setTimeout(() => {
+            controller.abort();
+          }, ANALYSIS_TIMEOUT_MS);
+
+          const response = await fetch(`${this.config.host}/api/generate`, {
+            method: 'POST',
+            headers: { 'Content-Type': 'application/json' },
+            body: JSON.stringify({
+              model: this.config.model,
+              prompt: userPrompt,
+              system: systemPrompt,
+              stream: false,
+              format: 'json',
+              options: { temperature },
+            }),
+            signal: controller.signal,
+          });
+
+          clearTimeout(timeoutId);
+
+          if (!response.ok) {
+            throw new Error(
+              `Ollama API failed: ${response.status.toString()} ${response.statusText}`,
+            );
+          }
+
+          const data = (await response.json()) as { response?: string };
+          if (typeof data.response !== 'string') {
+            throw new Error('Invalid response format from Ollama API');
+          }
+
+          // Parse response
+          let result: AnalysisResult;
+          logger.debug(
+            `Schema type: ${this.schemaType}, response length: ${String(data.response.length)}`,
+            'ollama',
           );
-        }
-
-        const data = (await response.json()) as { response?: string };
-
-        if (typeof data.response !== 'string') {
-          throw new Error('Invalid response format from Ollama API');
-        }
-
-        // Parse and validate the response based on schema type
-        if (this.schemaType === 'individual') {
-          return parseBatchIndividualResponse(data.response, date, prompts);
-        }
-        return parseResponse(data.response, date, undefined, prompts);
-      } catch (error) {
-        lastError = error instanceof Error ? error : new Error(String(error));
-
-        // Don't retry parse errors - they won't succeed on retry
-        if (
-          lastError.message.includes('parse') ||
-          lastError.message.includes('schema')
-        ) {
-          throw lastError;
-        }
-
-        // If this was the last attempt, throw the error
-        if (attempt === MAX_RETRIES) {
+          if (this.schemaType === 'individual') {
+            result = parseBatchIndividualResponse(data.response, date, prompts);
+          } else {
+            result = parseResponse(data.response, date, undefined, prompts);
+          }
+
+          // Semantic validation
+          const validation = validateSemantics(result, prompts);
+          if (!validation.valid) {
+            logger.debug(
+              'Semantic validation failed, attempting auto-correction',
+              'ollama',
+            );
+            result = autoCorrectResult(result, validation);
+          }
+
+          return result;
+        } catch (error) {
+          const err = error instanceof Error ? error : new Error(String(error));
+
+          // Parse errors: try lower temperature
+          if (err.message.includes('parse') || err.message.includes('schema')) {
+            lastParseError = err;
+            logger.debug(
+              `Parse failed at temp ${temperature.toString()}, trying lower`,
+              'ollama',
+            );
+            continue; // Try next temperature
+          }
+
+          // Network errors: break inner loop, retry outer
+          lastError = err;
           break;
         }
+      }
 
-        // Exponential backoff: 1s, 2s, 4s, ...
-        const delay = BASE_RETRY_DELAY_MS * Math.pow(2, attempt);
-        logger.debug(
-          `Retry attempt ${String(attempt + 1)}/${String(MAX_RETRIES)}, waiting ${String(delay)}ms`,
-          'ollama',
-        );
-        await sleep(delay);
+      // If we exhausted temperatures due to parse errors, throw
+      if (lastParseError && !lastError) {
+        throw lastParseError;
       }
+
+      // If this was the last network retry, break
+      if (attempt === MAX_RETRIES) break;
+
+      // Exponential backoff for network errors
+      const delay = BASE_RETRY_DELAY_MS * Math.pow(2, attempt);
+      logger.debug(
+        `Retry ${String(attempt + 1)}/${String(MAX_RETRIES)}, waiting ${String(delay)}ms`,
+        'ollama',
+      );
+      await sleep(delay);
     }
 
-    const attempts = String(MAX_RETRIES + 1);
     throw new Error(
-      `Ollama analysis failed after ${attempts} attempts: ${lastError?.message ?? 'Unknown error'}`,
+      `Ollama analysis failed after ${String(MAX_RETRIES + 1)} attempts: ${lastError?.message ?? lastParseError?.message ?? 'Unknown error'}`,
     );
   }
 }
 
-/**
- * Sleep utility for retry backoff.
- *
- * @param ms - Milliseconds to sleep
- * @returns Promise that resolves after the delay
- */
 function sleep(ms: number): Promise<void> {
   return new Promise((resolve) => setTimeout(resolve, ms));
 }
diff --git a/src/providers/schemas.test.ts b/src/providers/schemas.test.ts
index 849b2d0..1d23451 100644
--- a/src/providers/schemas.test.ts
+++ b/src/providers/schemas.test.ts
@@ -114,9 +114,9 @@ describe('schemas', () => {
     });
 
     it('should include examples', () => {
-      expect(SYSTEM_PROMPT_MINIMAL).toContain('Examples:');
+      expect(SYSTEM_PROMPT_MINIMAL).toContain('Examples');
       expect(SYSTEM_PROMPT_MINIMAL).toContain('Input:');
-      expect(SYSTEM_PROMPT_MINIMAL).toContain('Output:');
+      expect(SYSTEM_PROMPT_MINIMAL).toContain('->');
     });
 
     it('should specify score range', () => {
@@ -184,12 +184,11 @@ describe('schemas', () => {
   });
 
   describe('schema progression', () => {
-    it('should have minimal schema simpler than full schema', () => {
-      // Minimal should not require as many fields
-      const minimalFields = SYSTEM_PROMPT_MINIMAL.match(/"[^"]+"/g) ?? [];
-      const fullFields = SYSTEM_PROMPT_FULL.match(/"[^"]+"/g) ?? [];
-
-      expect(minimalFields.length).toBeLessThan(fullFields.length);
+    it('should have minimal schema shorter than full schema', () => {
+      // Minimal should be shorter in total length
+      expect(SYSTEM_PROMPT_MINIMAL.length).toBeLessThan(
+        SYSTEM_PROMPT_FULL.length,
+      );
     });
 
     it('should have consistent score range across all schemas', () => {
diff --git a/src/providers/schemas.ts b/src/providers/schemas.ts
index 88ebb88..bf7bc7c 100644
--- a/src/providers/schemas.ts
+++ b/src/providers/schemas.ts
@@ -133,32 +133,48 @@ export const ISSUE_TAXONOMY: IssueTaxonomy = {
  * Minimal system prompt for small models.
  * Returns only issue IDs and score - no examples or detailed metadata.
  */
-export const SYSTEM_PROMPT_MINIMAL = `You analyze prompts for quality issues.
+export const SYSTEM_PROMPT_MINIMAL = `You analyze coding prompts for quality issues.
 Respond with JSON only: {"issues": ["issue-id", ...], "score": 0-100}
 
 Valid issue IDs: vague, no-context, too-broad, no-goal, imperative, missing-technical-details, unclear-priorities, insufficient-constraints
 
 Issue definitions:
-- vague: Generic requests without specifics ("help", "fix", "improve")
-- no-context: Missing background info (uses "this", "it", "the bug" without context)
-- too-broad: Requests covering multiple unrelated topics
-- no-goal: Ambiguous success criteria or desired outcome
-- imperative: Commands without explanation or reasoning
-- missing-technical-details: No file paths, function names, or error messages
-- unclear-priorities: Multiple requests without ordering
-- insufficient-constraints: No requirements or edge cases mentioned
-
-Scoring: 0-100 (100=perfect, 90+=excellent, 70-89=good, 50-69=fair, <50=poor)
-
-Examples:
-Input: "Help me with code"
-Output: {"issues": ["vague", "no-context"], "score": 35}
-
-Input: "Debug this TypeScript function that returns undefined"
-Output: {"issues": ["missing-technical-details"], "score": 70}
-
-Input: "Debug calculateTotal() in utils.ts that returns undefined when called with empty array"
-Output: {"issues": [], "score": 90}`;
+- vague: Generic verbs without objects ("help", "fix", "improve" alone)
+- no-context: References without context ("this", "it", "the bug")
+- too-broad: Multiple unrelated tasks in one request
+- no-goal: No clear desired outcome or success criteria
+- imperative: Commands without explanation ("Add button")
+- missing-technical-details: No file paths, function names, errors
+- unclear-priorities: Multiple tasks without ordering
+- insufficient-constraints: No requirements or edge cases
+
+Scoring guide:
+- 90-100: Specific file/function, clear goal, actionable
+- 70-89: Good intent but missing some details
+- 50-69: Vague or missing context
+- 25-49: Multiple major issues
+- 0-24: Unusable (single words, no meaning)
+
+Examples (study the contrast):
+
+POOR (score 10-25):
+Input: "Help" -> {"issues": ["vague", "no-context", "no-goal"], "score": 10}
+Input: "Fix it" -> {"issues": ["vague", "no-context", "no-goal"], "score": 15}
+Input: "Debug" -> {"issues": ["vague", "no-context", "no-goal"], "score": 12}
+
+FAIR (score 45-60):
+Input: "Fix the login bug" -> {"issues": ["no-context", "missing-technical-details"], "score": 55}
+Input: "Add error handling" -> {"issues": ["no-context", "missing-technical-details"], "score": 52}
+Input: "Make it faster" -> {"issues": ["vague", "no-context"], "score": 48}
+
+GOOD (score 70-84):
+Input: "Fix the login bug where users cannot reset password" -> {"issues": ["missing-technical-details"], "score": 72}
+Input: "Add validation to signup form for email and password" -> {"issues": ["missing-technical-details"], "score": 78}
+
+EXCELLENT (score 85-100):
+Input: "Fix null pointer in auth.ts line 45 when user.email is undefined" -> {"issues": [], "score": 95}
+Input: "Add rate limiting to /api/login: 5 attempts per IP per minute" -> {"issues": [], "score": 93}
+Input: "Refactor calculateTotal() to use reduce, keep return type number" -> {"issues": [], "score": 88}`;
 
 /**
  * Simple system prompt for medium models.
diff --git a/src/types/index.ts b/src/types/index.ts
index f453209..79fbfbf 100644
--- a/src/types/index.ts
+++ b/src/types/index.ts
@@ -61,6 +61,7 @@ export type ExtractedPrompt = {
   readonly sessionId: string;
   readonly project: string;
   readonly date: string;
+  readonly isConfirmation?: boolean;
 };
 
 /**
@@ -188,23 +189,23 @@ export type BatchStrategy = {
  * Available batch strategies by model size.
  */
 export const BATCH_STRATEGIES: Record<BatchStrategyType, BatchStrategy> = {
-  // For models < 4GB
+  // Conservative: minimal schema, 1 prompt at a time
   micro: {
     maxTokensPerBatch: 500,
     maxPromptsPerBatch: 3,
-    description: 'For models < 4GB',
+    description: 'Conservative: individual schema, 1 prompt at a time',
   },
-  // For models 4-7GB
+  // Balanced: full schema, up to 10 prompts per batch
   small: {
     maxTokensPerBatch: 1_500,
     maxPromptsPerBatch: 10,
-    description: 'For models 4-7GB',
+    description: 'Balanced: full schema, up to 10 prompts per batch',
   },
-  // For models > 7GB
+  // Maximum: full schema, up to 50 prompts per batch
   standard: {
     maxTokensPerBatch: 3_000,
     maxPromptsPerBatch: 50,
-    description: 'For models > 7GB',
+    description: 'Maximum: full schema, up to 50 prompts per batch',
   },
 } as const;
 
@@ -316,7 +317,7 @@ export type EnvConfig = {
 export const ENV_DEFAULTS = {
   reminder: '7d',
   ollama: {
-    model: 'gemma3:4b',
+    model: 'gemma4:e4b',
     host: 'http://localhost:11434',
   },
   anthropic: {
diff --git a/tests/integration/api/library-usage.test.ts b/tests/integration/api/library-usage.test.ts
index 9285715..2a69da9 100644
--- a/tests/integration/api/library-usage.test.ts
+++ b/tests/integration/api/library-usage.test.ts
@@ -156,7 +156,7 @@ describe('Library API Usage', () => {
     it('should export ENV_DEFAULTS as constants', () => {
       expect(ENV_DEFAULTS).toBeDefined();
       expect(ENV_DEFAULTS.reminder).toBe('7d');
-      expect(ENV_DEFAULTS.ollama.model).toBe('gemma3:4b');
+      expect(ENV_DEFAULTS.ollama.model).toBe('gemma4:e4b');
       expect(ENV_DEFAULTS.ollama.host).toBe('http://localhost:11434');
       expect(ENV_DEFAULTS.anthropic.model).toBe('claude-3-5-haiku-latest');
       expect(ENV_DEFAULTS.google.model).toBe('gemini-2.0-flash-exp');
diff --git a/tests/integration/core/incremental-analysis.test.ts b/tests/integration/core/incremental-analysis.test.ts
index 0d6a31f..14abef6 100644
--- a/tests/integration/core/incremental-analysis.test.ts
+++ b/tests/integration/core/incremental-analysis.test.ts
@@ -188,7 +188,7 @@ describe('Incremental Analysis Integration', () => {
       // Check with different model
       const cacheCheck = await getPromptsWithCache(
         logResult.prompts,
-        'gemma3:4b', // Different model
+        'gemma4:e4b', // Different model
         'full',
       );
 
@@ -394,8 +394,8 @@ describe('Incremental Analysis Integration', () => {
       // Note: The function uses simple average, not weighted by prompt count
       expect(merged.stats.overallScore).toBe(6);
 
-      // Pattern frequency: average of 0.5 and 1.0 = 0.75, rounded to 1
-      expect(merged.patterns[0]?.frequency).toBe(1);
+      // Pattern frequency: sum of 0.5 and 1.0 = 1.5
+      expect(merged.patterns[0]?.frequency).toBe(1.5);
       // Severity should be the highest
       expect(merged.patterns[0]?.severity).toBe('high');
     });
diff --git a/tests/integration/core/performance.test.ts b/tests/integration/core/performance.test.ts
index 18fc6ed..240b882 100644
--- a/tests/integration/core/performance.test.ts
+++ b/tests/integration/core/performance.test.ts
@@ -43,7 +43,7 @@ describe('Performance - Incremental Analysis', () => {
       await populateResultsCache(resultsDir, testData, {
         date: '2025-01-20',
         project: 'perf-test',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -65,7 +65,7 @@ describe('Performance - Incremental Analysis', () => {
       const startTime = Date.now();
       const cacheCheck = await getPromptsWithCache(
         prompts,
-        'gemma3:4b',
+        'gemma4:e4b',
         'full',
       );
       const loadTime = Date.now() - startTime;
@@ -87,7 +87,7 @@ describe('Performance - Incremental Analysis', () => {
       await populateResultsCache(resultsDir, testData, {
         date: '2025-01-20',
         project: 'perf-test',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
       const populateTime = Date.now() - populateStart;
@@ -114,7 +114,7 @@ describe('Performance - Incremental Analysis', () => {
         const startTime = Date.now();
         const cacheCheck = await getPromptsWithCache(
           prompts,
-          'gemma3:4b',
+          'gemma4:e4b',
           'full',
         );
         const loadTime = Date.now() - startTime;
@@ -141,7 +141,7 @@ describe('Performance - Incremental Analysis', () => {
       const cachedData = generatePerformanceTestData(500);
       await populateResultsCache(resultsDir, cachedData, {
         date: '2025-01-20',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -173,7 +173,7 @@ describe('Performance - Incremental Analysis', () => {
       const startTime = Date.now();
       const cacheCheck = await getPromptsWithCache(
         mixedPrompts,
-        'gemma3:4b',
+        'gemma4:e4b',
         'full',
       );
       const loadTime = Date.now() - startTime;
@@ -205,7 +205,7 @@ describe('Performance - Incremental Analysis', () => {
       await populateResultsCache(resultsDir, testData, {
         date: '2025-01-20',
         project: 'perf-test',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -225,7 +225,7 @@ describe('Performance - Incremental Analysis', () => {
       // Load all cached results
       const cacheCheck = await getPromptsWithCache(
         prompts,
-        'gemma3:4b',
+        'gemma4:e4b',
         'full',
       );
 
@@ -265,7 +265,7 @@ describe('Performance - Incremental Analysis', () => {
       await populateResultsCache(resultsDir, testData, {
         date: '2025-01-20',
         project: 'perf-test',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -292,7 +292,7 @@ describe('Performance - Incremental Analysis', () => {
 
       // Perform multiple cache lookups
       for (let i = 0; i < 10; i++) {
-        await getPromptsWithCache(prompts, 'gemma3:4b', 'full');
+        await getPromptsWithCache(prompts, 'gemma4:e4b', 'full');
       }
 
       // Force garbage collection again
@@ -321,7 +321,7 @@ describe('Performance - Incremental Analysis', () => {
       await populateResultsCache(resultsDir, testData, {
         date: '2025-01-20',
         project: 'perf-test',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -344,7 +344,9 @@ describe('Performance - Incremental Analysis', () => {
       // Read all batches concurrently
       const startTime = Date.now();
       const results = await Promise.all(
-        batches.map((batch) => getPromptsWithCache(batch, 'gemma3:4b', 'full')),
+        batches.map((batch) =>
+          getPromptsWithCache(batch, 'gemma4:e4b', 'full'),
+        ),
       );
       const totalTime = Date.now() - startTime;
 
@@ -388,7 +390,7 @@ describe('Performance - Incremental Analysis', () => {
               date: '2025-01-20',
               project: 'perf-test',
               provider: 'test-provider',
-              model: 'gemma3:4b',
+              model: 'gemma4:e4b',
               schemaType: 'full',
             },
           ),
@@ -414,7 +416,7 @@ describe('Performance - Incremental Analysis', () => {
 
       const cacheCheck = await getPromptsWithCache(
         promptsToCheck,
-        'gemma3:4b',
+        'gemma4:e4b',
         'full',
       );
 
@@ -431,7 +433,7 @@ describe('Performance - Incremental Analysis', () => {
         const testData = generatePerformanceTestData(100, date);
         await populateResultsCache(resultsDir, testData, {
           date,
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         });
       }
@@ -454,7 +456,7 @@ describe('Performance - Incremental Analysis', () => {
       const startTime = Date.now();
       const cacheCheck = await getPromptsWithCache(
         prompts,
-        'gemma3:4b',
+        'gemma4:e4b',
         'full',
       );
       const loadTime = Date.now() - startTime;
@@ -479,7 +481,7 @@ describe('Performance - Incremental Analysis', () => {
         const testData = generatePerformanceTestData(50, date);
         await populateResultsCache(resultsDir, testData, {
           date,
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         });
       }
diff --git a/tests/integration/core/watch-mode-incremental.test.ts b/tests/integration/core/watch-mode-incremental.test.ts
index 4d590ea..be91eff 100644
--- a/tests/integration/core/watch-mode-incremental.test.ts
+++ b/tests/integration/core/watch-mode-incremental.test.ts
@@ -110,7 +110,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         {
           date: '2025-01-20',
           provider: 'test-provider',
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         },
       );
@@ -118,7 +118,7 @@ describe('Watch Mode - Incremental Analysis', () => {
       // Verify result was saved
       const cached = await getPromptResult('Fix authentication bug', {
         date: '2025-01-20',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -145,7 +145,7 @@ describe('Watch Mode - Incremental Analysis', () => {
 
       await populateResultsCache(resultsDir, cachedPrompts, {
         date: '2025-01-20',
-        model: 'gemma3:4b',
+        model: 'gemma4:e4b',
         schemaType: 'full',
       });
 
@@ -163,7 +163,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         // Check if prompt is cached
         const cached = await getPromptResult(event.prompt.content, {
           date: event.prompt.date,
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         });
 
@@ -222,7 +222,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         [{ content: 'Add tests', result: cachedResult }],
         {
           date: '2025-01-20',
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         },
       );
@@ -246,7 +246,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         const startTime = Date.now();
         const cached = await getPromptResult(event.prompt.content, {
           date: event.prompt.date,
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         });
         const retrievalTime = Date.now() - startTime;
@@ -293,7 +293,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         {
           date: '2025-01-20',
           project: 'test-project',
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         },
       );
@@ -312,7 +312,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         const cached = await getPromptResult(event.prompt.content, {
           date: event.prompt.date,
           project: event.prompt.project,
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         });
 
@@ -463,7 +463,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         try {
           await getPromptResult(event.prompt.content, {
             date: event.prompt.date,
-            model: 'gemma3:4b',
+            model: 'gemma4:e4b',
             schemaType: 'full',
           });
         } catch {
@@ -515,7 +515,7 @@ describe('Watch Mode - Incremental Analysis', () => {
         // Check cache
         const cached = await getPromptResult(event.prompt.content, {
           date: event.prompt.date,
-          model: 'gemma3:4b',
+          model: 'gemma4:e4b',
           schemaType: 'full',
         });
 
@@ -535,7 +535,7 @@ describe('Watch Mode - Incremental Analysis', () => {
             {
               date: event.prompt.date,
               provider: 'test-provider',
-              model: 'gemma3:4b',
+              model: 'gemma4:e4b',
               schemaType: 'full',
             },
           );
diff --git a/tests/integration/providers/index.test.ts b/tests/integration/providers/index.test.ts
index 2a32eda..1646fae 100644
--- a/tests/integration/providers/index.test.ts
+++ b/tests/integration/providers/index.test.ts
@@ -181,10 +181,9 @@ describe('Provider Integration - Ollama Provider', () => {
 
   it('should analyze prompts successfully', async () => {
     const config = {
-      model: 'llama3.2',
+      // Use standard-strategy model so auto-detection selects 'full' schema
+      model: 'llama3:70b',
       host: 'http://localhost:11434',
-      // Force batch schema to match the mock response format
-      schemaOverride: 'batch' as const,
     };
 
     const mockAnalysis = createMockAnalysis();
diff --git a/tests/unit/analytics/clustering.test.ts b/tests/unit/analytics/clustering.test.ts
index a782046..f89b6b3 100644
--- a/tests/unit/analytics/clustering.test.ts
+++ b/tests/unit/analytics/clustering.test.ts
@@ -31,8 +31,10 @@ describe('clustering', () => {
 
     for (const center of clusterCenters) {
       for (let i = 0; i < pointsPerCluster; i++) {
-        // Add small random variation around center
-        const point = center.map((val) => val + (Math.random() - 0.5) * 0.1);
+        // Add small deterministic variation around center
+        const point = center.map(
+          (val, d) => val + Math.sin(i * 7 + d * 3) * 0.01,
+        );
         embeddings.push(point);
       }
     }
@@ -42,10 +44,10 @@ describe('clustering', () => {
 
   describe('clusterPrompts', () => {
     it('should cluster embeddings into groups', () => {
-      // Create 3 distinct clusters
-      const cluster1Center = Array(10).fill(0.2);
-      const cluster2Center = Array(10).fill(0.5);
-      const cluster3Center = Array(10).fill(0.8);
+      // Create 3 distinct clusters with different directions (for cosine distance)
+      const cluster1Center = [1, 0, 0, 0, 0, 0, 0, 0, 0, 0];
+      const cluster2Center = [0, 0, 0, 1, 0, 0, 0, 0, 0, 0];
+      const cluster3Center = [0, 0, 0, 0, 0, 0, 0, 1, 0, 0];
 
       const embeddings = createClusterEmbeddings(
         [cluster1Center, cluster2Center, cluster3Center],
diff --git a/tests/unit/core/analyzer.test.ts b/tests/unit/core/analyzer.test.ts
index 2d1327d..a9e203f 100644
--- a/tests/unit/core/analyzer.test.ts
+++ b/tests/unit/core/analyzer.test.ts
@@ -285,7 +285,7 @@ describe('mergeBatchResults', () => {
     expect(ids).toHaveLength(3); // No duplicates
   });
 
-  it('should average frequencies for duplicate patterns', () => {
+  it('should sum frequencies for duplicate patterns across batches', () => {
     const results = [
       createResult({
         patterns: [createPattern('p1', { frequency: 4 })],
@@ -301,7 +301,7 @@ describe('mergeBatchResults', () => {
     });
 
     const pattern = merged.patterns.find((p) => p.id === 'p1');
-    expect(pattern?.frequency).toBe(5); // (4 + 6) / 2 = 5
+    expect(pattern?.frequency).toBe(10); // 4 + 6 = 10 (sum across batches)
   });
 
   it('should take max severity for duplicate patterns', () => {