fix: address PR review findings

- branch-analysis.js: fix '=' .repeat(70) syntax error → '='.repeat(70)
- package.json: remove setup script pointing to deleted download-model.sh
- config/index.js: CLICKHOUSE_PASSWORD required only in production (dev/test warning)
- generator.js: pass config.model.revision to pipelineFn for reproducibility
- server.js, Dockerfile: update version 2.0.0 → 2.1.0

Author: tbartel74

services/semantic-service/Dockerfile

@@ -14,7 +14,7 @@ RUN npm install -g npm@latest
 # Labels
 LABEL org.opencontainers.image.title="Semantic Service"
 LABEL org.opencontainers.image.description="Semantic similarity detection using E5 multilingual embeddings"
-LABEL org.opencontainers.image.version="2.0.0"
+LABEL org.opencontainers.image.version="2.1.0"
 LABEL org.opencontainers.image.vendor="Vigil Guard"
 
 # Security: Run as non-root user

services/semantic-service/package.json

@@ -12,7 +12,7 @@
     "test:coverage": "vitest run --coverage",
     "test:golden": "RUN_GOLDEN_TESTS=1 vitest run tests/golden-dataset/",
     "lint": "eslint src/ tests/",
-    "setup": "./scripts/download-model.sh",
+    "setup": "echo 'Embeddings pre-generated in data/datasets/'",
     "embeddings:generate": "python3 scripts/generate-embeddings.py",
     "embeddings:import": "node scripts/import-embeddings.js",
     "embeddings:manage": "node scripts/embedding-manager.js"

services/semantic-service/src/config/index.js

@@ -126,11 +126,14 @@ const config = {
 function validateConfig() {
     const errors = [];
 
-    // ClickHouse password required to avoid empty-auth startup
-    if (!config.clickhouse.password) {
+    // ClickHouse password required in production
+    if (!config.clickhouse.password && config.server.env === 'production') {
         console.error('FATAL: CLICKHOUSE_PASSWORD not set');
         process.exit(1);
     }
+    if (!config.clickhouse.password && config.server.env !== 'production') {
+        console.warn('WARN: CLICKHOUSE_PASSWORD not set (ok for dev/test)');
+    }
 
     if (config.model.dimension !== 384) {
         errors.push('Model dimension must be 384 for E5-small');

services/semantic-service/src/embedding/generator.js

@@ -40,8 +40,8 @@ async function initialize() {
     // Transformers.js will first check cacheDir for the model, then fetch if not found
     extractor = await pipelineFn('feature-extraction', modelId, {
         quantized: true,
-        // If locally available, prevent network fetch
-        local_files_only: isLocallyAvailable
+        local_files_only: isLocallyAvailable,
+        revision: config.model.revision
     });
 
     ready = true;

services/semantic-service/src/server.js

@@ -450,7 +450,7 @@ app.get('/metrics', async (req, res) => {
 app.get('/', (req, res) => {
     res.json({
         service: 'semantic-service',
-        version: '2.0.0',
+        version: '2.1.0',
         branch: config.branch,
         description: 'Semantic similarity detection using E5 multilingual embeddings',
         endpoints: {

services/workflow/tests/branch-analysis.js

@@ -152,9 +152,9 @@ function analyzeBranchContribution(event, expected) {
  * Main analysis
  */
 async function main() {
-    console.log('=' .repeat(70));
+    console.log('='.repeat(70));
     console.log('Branch Specialization Analysis');
-    console.log('=' .repeat(70));
+    console.log('='.repeat(70));
 
     // Load fixtures
     const allPrompts = loadFixtures();