diff --git a/.github/workflows/npm-release.yml b/.github/workflows/npm-release.yml
index 9979ba3..ddf2c0e 100644
--- a/.github/workflows/npm-release.yml
+++ b/.github/workflows/npm-release.yml
@@ -6,7 +6,7 @@ on:
       version:
         description: Lockstep version for all release packages
         required: true
-        default: 0.3.1
+        default: 0.3.2
         type: string
       publish_to_registry:
         description: Publish the fully gated candidate to npm after preflight
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6991bc0..e29e4ce 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,31 @@
 
 This file records user-visible changes to `light-ocr`. Published artifact details and immutable hashes remain in [`docs/releases/`](docs/releases/).
 
+## [0.3.2] - 2026-07-22
+
+### Added
+
+- Added the `light-ocr` CLI bin with three subcommands: `recognize` (default OCR), `detect` (detection-only, no recognition), and `info` (engine/version diagnostics). The CLI is pure JavaScript (CJS, zero-dependency) and ships as a `bin` entry in the facade package.
+- Added a versioned `DocumentResult` envelope (`schemaVersion: 1`) wrapping `OcrResult` with stable line IDs (`L0`, `L1`, …) and detection IDs (`D0`, `D1`, …) for future Layout region association.
+- Added `--schema-version 1` to request an exact output schema; unsupported versions return `invalid_argument`.
+- Added `--region x,y,w,h` ROI support: pageSpace axis-aligned rectangle, adapter-layer pixel crop after EXIF correction, box coordinates offset back to full pageSpace. Out-of-bounds or partially-intersecting regions return `invalid_argument` without clamping.
+- Added EXIF orientation correction for JPEG images: a self-contained C++ EXIF parser reads the orientation tag and applies the pixel transform before recognition. `--no-exif` disables correction; `appliedTransforms` records the applied state.
+- Added `detect` subcommand support in Core: `Engine::detect()` reuses the detection stage and skips recognition. `DetectionResult` carries boxes (Quad + score), image dimensions, model bundle ID, and timing.
+- Added `applyExif` and `region` options to `RecognizeOptions` (Node API + TypeScript types).
+- Added an Agent Skill at `.agents/skills/local-ocr/SKILL.md` with scenario-driven workflows, decision flow, and exit code reference.
+- Added stable exit codes (64–72) mapped to `OcrErrorCode`, with flag validation ordered before input reading so parameter errors surface first.
+- Added a Windows delay-load hook (`win_delay_load_hook.cpp`) so the native addon loads under Electron and other renamed Node-API hosts, not just `node.exe`.
+
+### Changed
+
+- The `detect` subcommand does not expose `--format`; its output is always structured JSON with `structure: "detect"` and `detections[]` instead of `lines[]`.
+- `info --model-info` and `info --version` are mutually exclusive (version triple is a subset of EngineInfo).
+- CLI flag surface restricted to Agent/user-facing options; provider-internal fields (`sessionFallback`, `cpuPartition`, `precision`, `detectionStrategy`, `maxSide`, `includeDiagnostics`) are not exposed as CLI flags.
+
+### Notes
+
+- This is the first release with the N1 CLI entry point (roadmap §5). The CLI, EXIF, ROI, and detect features are newly implemented; Agent task eval and npm install smoke validation follow per roadmap §5.7.
+
 ## [0.3.1] - 2026-07-21
 
 ### Added
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 2fd5a70..882a76a 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -7,7 +7,7 @@ if(APPLE AND (NOT DEFINED CMAKE_OSX_DEPLOYMENT_TARGET OR
     "Minimum macOS deployment target" FORCE)
 endif()
 
-project(light_ocr VERSION 0.3.1 LANGUAGES CXX)
+project(light_ocr VERSION 0.3.2 LANGUAGES CXX)
 
 if(APPLE)
   enable_language(OBJCXX)
diff --git a/bindings/node/README.md b/bindings/node/README.md
index 4757a44..5c22329 100644
--- a/bindings/node/README.md
+++ b/bindings/node/README.md
@@ -1,179 +1,237 @@
-# light-ocr Node-API adapter
+# light-ocr
 
-状态：`@arcships/light-ocr@0.3.0` 已发布并提升为 npm `latest`。默认 `createEngine()` 使用 descriptor-driven Auto；macOS arm64 可用 Apple/Core ML，Linux x64 glibc/Vulkan 与 Windows x64/D3D12 可用 official Native WebGPU Plugin EP，macOS x64 保持 CPU-only。
+Local OCR for Node.js and Agents. Offline, no Python, no network calls. Returns text with coordinates and confidence.
 
-推荐直接安装公开 package：
+## Install
 
 ```bash
 npm install @arcships/light-ocr
 ```
 
-安装会带上 PP-OCRv6 Small 模型与当前平台的预编译 native package；`createEngine()` 默认无需 `bundlePath`。以下本地构建说明用于修改或调试 adapter。
+The install includes the PP-OCRv6 Small model and a prebuilt native runtime for your platform. No post-install scripts, no downloads — `createEngine()` works out of the box.
 
-## 能力边界
+## CLI
 
-- 原始 Node-API C API，编译为 `NAPI_VERSION=8`，不依赖 `node-addon-api`。
-- `createEngine()`、`recognize()`、`close()` 全部返回 Promise。
-- 每个 engine 一条专用 C++ worker thread 和一个有界 FIFO；推理不占 JavaScript 线程或 libuv 共享线程池。
-- `recognize()` 接受 `Uint8Array` raw pixels：`gray8`、`rgb8`、`bgr8`、`rgba8`。
-- `recognizeEncoded()` 接受内存中的 JPEG/PNG `Uint8Array`；格式自动检测，解码在 engine worker 中执行。
-- `recognize()` 返回前同步复制本次调用实际需要的像素范围；调用返回后可以立即修改或复用原 Buffer。
-- 支持 `AbortSignal` 协作式取消：queued 请求会从队列移除；running 请求立即拒绝 public Promise，但 Core 会安全运行到返回并丢弃结果。
-- native addon 只接收现有绝对 bundle 目录。当前源码开发调用显式传 `bundlePath`；发布后的 facade 默认使用随 npm 安装的 model package 路径。
-- 产品 engine 默认报告 `detectionStrategy: 'bounded'`、`detectionMaxSide: 960` 和 `defaultRecognitionBatchSize: 1`。0.3.0 可通过 `detection: {strategy: 'tiled'}` 显式选择 `tiled-v1`；`upstreamExact` 只用于上游对照，单次 `recognize({detectionMaxSide})` 只能继续降低 bounded engine 的 side。
-- `createEngine({execution})` 接受 `auto`、`cpu`、`apple` 与已交付平台支持的 `webgpu`。macOS 15+ arm64 默认开放：Apple Silicon interactive 使用 FP16 ANE + 宽文本 FP16 GPU，strict 使用全 GPU；macOS x64 package 保持 CPU-only。WebGPU 使用 ORT Core 1.24.4 + official plugin 0.1.0，Linux 为 Vulkan、Windows 为 D3D12；`0.3.0` 公共 WebGPU profile 只接受 `precision: 'auto' | 'fp32'`，FP16 仅用于 Apple provider。当前 WebGPU 模型需要 `Concat/Gather/Slice` 三类有界 CPU partition，因而 `cpuPartition: 'forbid'` 会稳定 fail-closed。只有 Auto 可在创建期按 descriptor 锁定的 typed failure 继续候选；显式 provider 不回退，旧 `sessionFallback: 'cpu'` 返回 `invalid_argument`。`engine.info.execution.sessions` 报告每个模型的实际 provider chain、precision、adapter、runtime/provider/qualification identity，selection trace 则报告 Auto 的每次创建尝试。
+The `light-ocr` command is available after install.
 
-## `0.3.0` 加速证据
+### Quick start
 
-| Provider 与记录设备 | P50 结果 | 质量与 Gate |
-| --- | ---: | --- |
-| Apple/Core ML，Apple M4 Max | `HELLO 123` 2.30×；XFUND 2.85× | 14 fixtures 通过 CPU parity 阈值 |
-| WebGPU/Vulkan，NVIDIA RTX 5060 Ti | 14-fixture 聚合 5.70×；单项 3.47×–9.30× | 14/14 与 CPU FP32 一致；164/164 |
-| WebGPU/D3D12，AMD Radeon 780M | 14-fixture 聚合 2.44×；单项 1.28×–2.98× | 14/14 与 CPU FP32 一致；164/164 |
+```bash
+# Recognize text in an image (default action)
+light-ocr image.png --format json
 
-这些结果都是表中设备上的同机 CPU 对照。WebGPU 聚合值为 14 个 fixture 的 CPU P50 总和除以 WebGPU P50 总和，不外推到其他 GPU/driver。
+# Just text, no coordinates
+light-ocr image.png --format text
 
-不支持 WebP、GIF、PDF、EXIF orientation 自动旋转、zero-copy/transfer、运行中 inference 硬中断或 Bun。Windows addon 使用 `node.exe` delay-load hook，可由 Node 或重命名宿主（包括 Electron）解析 Node-API symbols；Electron 的完整版本、ASAR、worker 与 lifecycle 支持矩阵仍按独立 Gate 管理。详细契约见 [Node-API 设计](../../docs/napi-design.md)。
+# Engine info
+light-ocr info --version
+light-ocr info --model-info
+```
 
-## 本地构建
+### Subcommands
 
-先按仓库根目录文档准备锁定的依赖缓存和 PP-OCRv6 bundle。Node headers 必须显式传给 CMake：
+| Command | Purpose |
+| --- | --- |
+| `recognize <path> [flags]` | Full OCR: detection + recognition (default) |
+| `detect <path> [flags]` | Detection only: text region boxes, no recognition |
+| `info --model-info \| --version` | Engine/version diagnostics, no image read |
 
-```bash
-NODE_INCLUDE_DIR="$(node -p \
-  "require('node:path').resolve(require('node:path').dirname(process.execPath), '../include/node')")"
-
-cmake -S . -B build-node \
-  -DCMAKE_BUILD_TYPE=Debug \
-  -DLIGHT_OCR_DEPENDENCY_CACHE_DIR="$PWD/.cache/dependencies" \
-  -DLIGHT_OCR_BUILD_NODE=ON \
-  -DLIGHT_OCR_BUILD_TESTS=ON \
-  -DLIGHT_OCR_BUILD_TOOLS=OFF \
-  -DLIGHT_OCR_NODE_INCLUDE_DIR="$NODE_INCLUDE_DIR" \
-  -DLIGHT_OCR_NODE_EXECUTABLE="$(command -v node)"
+`light-ocr image.png` without a subcommand is implicit `recognize`.
 
-cmake --build build-node --target light_ocr_node --parallel
-```
+### Common flags
+
+| Flag | Values | Description |
+| --- | --- | --- |
+| `--format` | `json` \| `jsonl` \| `text` | Output format (default: `json`) |
+| `--region` | `x,y,w,h` | Restrict recognition to a pageSpace rectangle |
+| `--provider` | `auto` \| `cpu` \| `apple` \| `webgpu` | Execution provider (default: `auto`) |
+| `--no-exif` | — | Disable EXIF orientation correction |
+| `--schema-version` | `1` | Request exact output schema |
+| `--quiet` | — | Suppress non-error stderr |
 
-macOS/Linux 的链接产物在 `build-node/bin/light_ocr_node.node`；可加载的完整开发 runtime 会连同 descriptor 与锁定的 ONNX Runtime 一起放在 `build-node/node-runtime/native/`。Windows 还需通过 `LIGHT_OCR_NODE_LIBRARY` 指定当前架构的 `node.lib`，staged runtime 位于 `build-node/node-runtime/Release/native/`。Windows 构建会把 delay-load hook 直接链接进 `.node`，并在产物生成后验证 `node.exe` 只存在于 PE delay-import directory；这样同一个 N-API addon 才能由 Node 和 Electron 的当前可执行文件提供 symbols。
+`detect` does not accept `--format` (output is always JSON with `detections[]`). `detect --crop` attaches a PNG crop per detection.
 
-WebGPU 构建必须先用 [`tools/webgpu/build_runtime.py`](../../tools/webgpu/README.md) 生成并验证目标 SDK，再增加：
+### Scenarios
 
+**Screenshot with small text:**
 ```bash
-cmake -S . -B build-node-webgpu -G Ninja \
-  -DCMAKE_BUILD_TYPE=Release \
-  -DLIGHT_OCR_DEPENDENCY_CACHE_DIR="$PWD/.cache/dependencies" \
-  -DLIGHT_OCR_ONNXRUNTIME_FLAVOR=webgpu \
-  -DLIGHT_OCR_WEBGPU_SDK_DIR="$PWD/dist/webgpu-sdk/linux-x64" \
-  -DLIGHT_OCR_WEBGPU_QUALIFICATION_BUILD=ON \
-  -DLIGHT_OCR_BUILD_NODE=ON \
-  -DLIGHT_OCR_NODE_INCLUDE_DIR="$NODE_INCLUDE_DIR" \
-  -DLIGHT_OCR_NODE_EXECUTABLE="$(command -v node)"
+light-ocr screenshot.png --format json
+# Low confidence? Re-run on a specific region:
+light-ocr recognize screenshot.png --region 100,80,640,320 --format json
 ```
 
-qualification staged payload 在 Linux 包含 core + plugin，在 Windows 还包含 `dxcompiler.dll` 与 `dxil.dll`。schema 2 descriptor 精确声明所有 native 文件及 SHA-256；loader 从 sterile cwd 验证完整 inventory 后才加载 addon，Core 在注册 plugin 前再次复核 provider library。普通 release 构建不接受 pending qualification lock。
+**Form/receipt field extraction (detect-then-recognize):**
+```bash
+light-ocr detect receipt.png                    # find where text is
+light-ocr recognize receipt.png --region 50,200,300,80 --format json  # read that area
+```
 
-开发期用 `LIGHT_OCR_NODE_BINARY` 指向刚构建的 addon：
+**stdin / pipe:**
+```bash
+cat image.png | light-ocr recognize --stdin --type image/png --format json
+```
 
+**Batch via shell:**
 ```bash
-export LIGHT_OCR_NODE_BINARY="$PWD/build-node/node-runtime/native/light_ocr_node.node"
-export LIGHT_OCR_RUNTIME_DESCRIPTOR="$PWD/build-node/node-runtime/native/runtime-descriptor.json"
-export LIGHT_OCR_MODEL_BUNDLE="$PWD/models/generated/ppocrv6-small-webgpu-20260719.1"
-export LIGHT_OCR_APPLE_MODEL_BUNDLE="$PWD/models/generated/ppocrv6-small-native-20260719.1"
+for f in *.png; do light-ocr recognize "$f" --format jsonl; done
+```
 
-node --test --test-concurrency=1 bindings/node/test/adapter.test.cjs
-# 或：ctest --test-dir build-node -R '^light_ocr_node_tests$' --output-on-failure
+### Output schema
+
+```json
+{
+  "schemaVersion": 1,
+  "source": { "kind": "image", "mediaType": "image/png", "identity": {}, "appliedTransforms": {} },
+  "pages": [{
+    "index": 0,
+    "width": 640,
+    "height": 480,
+    "coordinateSpace": "pageSpace",
+    "structure": "ocr-order",
+    "lines": [{ "id": "L0", "text": "HELLO", "confidence": 0.99, "box": [{"x":0,"y":0},{"x":100,"y":0},{"x":100,"y":30},{"x":0,"y":30}] }]
+  }]
+}
 ```
 
-当前源码加载器不会搜索任意 cwd。开发构建优先接受 `LIGHT_OCR_NODE_BINARY`，发布构建则按 `process.platform`、`process.arch` 和 Linux libc 固定选择六个 `@arcships/light-ocr-<platform>` optional packages。facade 通过 model package 导出的 manifest 定位默认 bundle，并在进入 native addon 前核对 bundle ID。详见 [npm package 设计](../../docs/npm-packaging.md)。
+- `box` is 4 points in `pageSpace` (top-left origin, x right, y down, post-EXIF pixels)
+- `detect` replaces `lines` with `detections[]` (`{ id, score, box }`) and sets `structure: "detect"`
+- `--format text`: recognized text only, one line per line
+- `--format jsonl`: one page record per line (for streaming/batch)
+
+### Exit codes
 
-## 使用
+| Code | Meaning |
+| --- | --- |
+| 0 | Success |
+| 64 | Usage error |
+| 65 | Invalid argument (bad region, unsupported format/schema) |
+| 66 | Invalid image |
+| 67 | Unsupported capability |
+| 68 | Model/bundle error |
+| 69 | Resource limit exceeded |
+| 70 | Environment/package failure |
+| 71 | Inference failure |
+| 72 | Internal error |
 
-发布后的目标用法不需要模型路径：
+## Node API
 
 ```js
-const { createEngine, OcrError } = require('@arcships/light-ocr');
+const { createEngine } = require('@arcships/light-ocr');
+
+const engine = await createEngine();
 
+const result = await engine.recognizeEncoded(imageBytes);
+console.log(result.lines[0].text);        // "HELLO"
+console.log(result.lines[0].confidence);   // 0.99
+console.log(result.lines[0].box);           // [{x,y}, {x,y}, {x,y}, {x,y}]
+
+await engine.close();
+```
+
+### Detect (detection only, no recognition)
+
+```js
+const engine = await createEngine();
+const result = await engine.detect(imageBytes);
+console.log(result.boxes[0].score);  // 0.95
+console.log(result.boxes[0].box);    // [{x,y}, {x,y}, {x,y}, {x,y}]
+await engine.close();
+```
+
+### With options
+
+```js
 const engine = await createEngine({
+  execution: { provider: 'auto' },
   queueCapacity: 4,
-  execution: {
-    provider: 'webgpu',
-    precision: 'fp32',
-    cpuPartition: 'allow',
-    sessionFallback: 'error',
-  },
+});
+
+const result = await engine.recognizeEncoded(imageBytes, {
+  applyExif: true,       // EXIF orientation correction (default: true)
+  region: { x: 100, y: 80, width: 640, height: 320 },  // ROI
+  signal: controller.signal,
 });
 
 console.log(engine.info.execution.sessions.detection.actualProviderChain);
-console.log(engine.info.execution.sessions.detection.deviceValidated);
+await engine.close();
 ```
 
-当前源码开发用法仍需显式 bundle：
-
-CommonJS：
+### ESM
 
 ```js
-const { createEngine, OcrError } = require('./bindings/node/js/index.cjs');
-
-async function main() {
-  const engine = await createEngine({
-    bundlePath: '/absolute/path/to/ppocrv6-small-bundle',
-    queueCapacity: 4,
-    maxPendingInputBytes: 256 * 1024 * 1024,
-  });
-
-  try {
-    const controller = new AbortController();
-    const result = await engine.recognize(
-      {
-        data: bgrPixels,
-        width,
-        height,
-        stride,
-        pixelFormat: 'bgr8',
-      },
-      {
-        includeDiagnostics: true,
-        signal: controller.signal,
-      },
-    );
-    const encodedResult = await engine.recognizeEncoded(
-      await require('node:fs/promises').readFile('/absolute/path/to/image.jpg'),
-      { signal: controller.signal },
-    );
-    console.log(result.lines);
-    console.log(encodedResult.lines);
-  } catch (error) {
-    if (error instanceof OcrError) console.error(error.code, error.message, error.detail);
-    else throw error; // 包括调用方提供的 AbortSignal.reason
-  } finally {
-    await engine.close();
-  }
-}
-
-main().catch(console.error);
+import { createEngine } from '@arcships/light-ocr';
 ```
 
-发布包的 ESM 使用 `import { createEngine, OcrError } from '@arcships/light-ocr'`；当前源码直接加载 `./bindings/node/js/index.mjs`。完整 TypeScript 类型位于 [index.d.ts](js/index.d.ts)，目标 package API 见 [Node-API 设计](../../docs/napi-design.md)。
+TypeScript types are included in [`index.d.ts`](js/index.d.ts).
+
+## Capabilities
+
+- **Formats**: JPEG, PNG (memory input via `Uint8Array`)
+- **EXIF orientation**: JPEG orientation tags 1–8 automatically corrected
+- **ROI**: pageSpace axis-aligned rectangle, coordinates offset back to full page
+- **Providers**: CPU (all platforms), Apple Core ML (macOS arm64), WebGPU (Linux x64, Windows x64)
+- **Offline**: no network at install or runtime; model bundled in npm package
+- **Concurrency**: one engine = one worker thread, bounded FIFO queue, AbortSignal support
+- **Platforms**: macOS arm64/x64, Linux x64/arm64 (glibc), Windows x64/arm64
+- **Node.js**: 22, 24
 
-## 背压与取消
+### Not supported
 
-`queueCapacity` 统计同一 engine 的 running + queued 请求；`maxPendingInputBytes` 统计尚未被 Core 消费完的输入快照。任一临时不足都会以 `OcrError`/`queue_full` 拒绝，不会隐式等待或无界增长。单个输入自身超过 byte budget 时返回 `resource_limit_exceeded`。
+- WebP, GIF, PDF, TIFF
+- Character-level coordinates (recognition is line-level)
+- Multi-engine fan-out as default architecture
+- Bun (Node-API lifecycle not fully verified)
 
-AbortSignal 的 public Promise 以最先观察到的终态为准：
+## Acceleration
 
-- 调用前已 aborted：不 admission，不复制 pixels，按原始 `signal.reason` 拒绝。
-- queued：立即释放 request slot、snapshot 和 byte budget，永不进入 Core。
-- running：立即按 reason 拒绝；底层 inference 继续，engine 在它返回前仍占用执行槽。
+`provider: 'auto'` (default) selects the best available backend:
 
-这不是硬超时。需要强制终止不可信或严格 deadline 的任务时，应使用独立进程作为隔离边界。
+| Platform | Auto candidate | Notes |
+| --- | --- | --- |
+| macOS arm64 | Apple Core ML → CPU | FP16 detection + recognition via Neural Engine / GPU |
+| macOS x64 | CPU | Core ML parity not met on Intel |
+| Linux x64 | WebGPU → CPU | FP32, bounded CPU partition for Concat/Gather/Slice |
+| Windows x64 | WebGPU → CPU | FP32, D3D12 |
+| Linux arm64 | CPU | WebGPU plugin has no arm64 binary |
+| Windows arm64 | CPU | WebGPU not yet qualified |
 
-## 生命周期
+Use `light-ocr info --model-info` to inspect the actual provider chain, device, and qualification status.
 
-`close()` 幂等，并按 FIFO drain 已接收的请求；close 开始后不再接收新请求。建议始终在 `finally` 中 await close。忘记 close 时，wrapper finalizer 只发出非阻塞关闭请求；Node Environment teardown 会停止 completion transport、丢弃未开始请求并等待正在运行的 Core 调用安全结束。
+## Build from source
 
-每个 `worker_threads` Environment 有独立 dispatcher、engine registry 和 request ID 空间；engine 对象不能跨 Environment 传递。
+See [CONTRIBUTING.md](../../CONTRIBUTING.md) for prerequisites and the full build guide.
+
+```bash
+# Clone and initialize
+git clone https://github.com/arcships/light-ocr.git
+cd light-ocr
+
+# Build C++ core
+cmake -S . -B build-ci -G Ninja -DCMAKE_BUILD_TYPE=Release \
+  -DLIGHT_OCR_DEPENDENCY_CACHE_DIR="$PWD/.cache/dependencies"
+cmake --build build-ci --parallel
+ctest --test-dir build-ci --output-on-failure
+
+# Build Node addon
+cd bindings/node
+NODE_INCLUDE_DIR="$(node -p "require('node:path').resolve(require('node:path').dirname(process.execPath), '../include/node')")"
+
+cmake -S ../.. -B build-node -G Ninja -DCMAKE_BUILD_TYPE=Release \
+  -DLIGHT_OCR_DEPENDENCY_CACHE_DIR="$PWD/../../.cache/dependencies" \
+  -DLIGHT_OCR_BUILD_NODE=ON -DLIGHT_OCR_BUILD_TESTS=ON -DLIGHT_OCR_BUILD_TOOLS=OFF \
+  -DLIGHT_OCR_NODE_INCLUDE_DIR="$NODE_INCLUDE_DIR" \
+  -DLIGHT_OCR_NODE_EXECUTABLE="$(command -v node)"
+cmake --build build-node --target light_ocr_node --parallel
+```
+
+Windows builds additionally require `LIGHT_OCR_NODE_LIBRARY` pointing to `node.lib`. The build links a delay-load hook so the addon works under Electron as well as Node.
+
+Run tests:
+
+```bash
+node --test --test-concurrency=1 bindings/node/test/*.test.cjs
+```
 
-## Sanitizer 注意事项
+## License
 
-addon 和它所链接的 Core 可以用项目的 ASan/UBSan 或 TSan 选项编译，但测试进程也必须是与该 sanitizer 兼容的 Node host。macOS 官方签名 Node 可能清除 `DYLD_INSERT_LIBRARIES`，从而以“interceptors loaded too late”在加载 addon 前退出；release gate 应使用专门的 instrumented Node/runner，不能把 Core executable 的 sanitizer 启动命令原样套给 `.node`。
+Apache-2.0. The bundled PP-OCRv6 model inherits PaddleOCR's license terms.
diff --git a/bindings/node/bin/light-ocr.cjs b/bindings/node/bin/light-ocr.cjs
index d212bba..2f16057 100644
--- a/bindings/node/bin/light-ocr.cjs
+++ b/bindings/node/bin/light-ocr.cjs
@@ -18,7 +18,7 @@ const { createEngine, OcrError } = require('../js/index.cjs');
 const { parseExifOrientation } = require('../js/exif.cjs');
 
 const PKG_VERSION = require('../package.json').version;
-const CORE_VERSION = '0.3.1';
+const CORE_VERSION = '0.3.2';
 
 const SUBCOMMANDS = new Set(['recognize', 'detect', 'info']);
 const EXIT = {
diff --git a/bindings/node/package.json b/bindings/node/package.json
index 4d8fa78..3b71a1f 100644
--- a/bindings/node/package.json
+++ b/bindings/node/package.json
@@ -1,6 +1,6 @@
 {
   "name": "@arcships/light-ocr",
-  "version": "0.3.1",
+  "version": "0.3.2",
   "private": true,
   "description": "Node-API adapter for the light-ocr C++ core",
   "license": "Apache-2.0",