fix(reflection): improve plan mode and route feedback

.eval-tmp/opencode.json

@@ -0,0 +1,4 @@
+{
+  "$schema": "https://opencode.ai/config.json",
+  "model": "github-copilot/gpt-4o"
+}

.github/workflows/deploy-supabase.yml

@@ -0,0 +1,55 @@
+name: Deploy Supabase
+
+on:
+  push:
+    branches:
+      - main
+      - master
+    paths:
+      - 'supabase/**'
+      - '.github/workflows/deploy-supabase.yml'
+  workflow_dispatch:
+    inputs:
+      deploy_target:
+        description: 'What to deploy'
+        required: true
+        default: 'all'
+        type: choice
+        options:
+          - all
+          - functions
+          - migrations
+
+jobs:
+  deploy:
+    name: Deploy to Supabase
+    runs-on: ubuntu-latest
+
+    env:
+      SUPABASE_ACCESS_TOKEN: ${{ secrets.SUPABASE_ACCESS_TOKEN }}
+      SUPABASE_PROJECT_REF: ${{ secrets.SUPABASE_PROJECT_REF || 'slqxwymujuoipyiqscrl' }}
+      SUPABASE_DB_PASSWORD: ${{ secrets.SUPABASE_DB_PASSWORD }}
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup Supabase CLI
+        uses: supabase/setup-cli@v1
+        with:
+          version: latest
+
+      - name: Verify Supabase CLI
+        run: supabase --version
+
+      - name: Deploy (push event)
+        if: github.event_name == 'push'
+        run: ./scripts/deploy-supabase.sh functions
+
+      - name: Deploy (manual trigger)
+        if: github.event_name == 'workflow_dispatch'
+        run: ./scripts/deploy-supabase.sh ${{ inputs.deploy_target }}
+
+      - name: List deployed functions
+        run: |
+          supabase functions list --project-ref "$SUPABASE_PROJECT_REF"

.github/workflows/evals.yml

@@ -0,0 +1,155 @@
+name: Prompt Evaluations
+
+on:
+  # Run on PRs to validate prompt quality
+  pull_request:
+    branches:
+      - main
+      - master
+    paths:
+      - 'reflection.ts'
+      - 'evals/**'
+  # Manual trigger for full evaluation
+  workflow_dispatch:
+    inputs:
+      eval_type:
+        description: 'Which evaluation to run'
+        required: true
+        default: 'all'
+        type: choice
+        options:
+          - all
+          - judge
+          - stuck
+          - compression
+
+permissions:
+  contents: read
+  pull-requests: write
+
+jobs:
+  evaluate:
+    name: Run Prompt Evaluations
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+          cache: 'npm'
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Create results directory
+        run: mkdir -p evals/results
+
+      - name: Run Judge Evaluation
+        if: ${{ github.event.inputs.eval_type == 'all' || github.event.inputs.eval_type == 'judge' || github.event_name == 'pull_request' }}
+        env:
+          AZURE_OPENAI_API_KEY: ${{ secrets.AZURE_OPENAI_API_KEY }}
+          AZURE_OPENAI_ENDPOINT: ${{ secrets.AZURE_OPENAI_ENDPOINT }}
+        run: npm run eval:judge -- --no-progress-bar -o evals/results/judge-results.json
+        continue-on-error: true
+
+      - name: Run Stuck Detection Evaluation
+        if: ${{ github.event.inputs.eval_type == 'all' || github.event.inputs.eval_type == 'stuck' }}
+        env:
+          AZURE_OPENAI_API_KEY: ${{ secrets.AZURE_OPENAI_API_KEY }}
+          AZURE_OPENAI_ENDPOINT: ${{ secrets.AZURE_OPENAI_ENDPOINT }}
+        run: npm run eval:stuck -- --no-progress-bar -o evals/results/stuck-results.json
+        continue-on-error: true
+
+      - name: Run Post-Compression Evaluation
+        if: ${{ github.event.inputs.eval_type == 'all' || github.event.inputs.eval_type == 'compression' }}
+        env:
+          AZURE_OPENAI_API_KEY: ${{ secrets.AZURE_OPENAI_API_KEY }}
+          AZURE_OPENAI_ENDPOINT: ${{ secrets.AZURE_OPENAI_ENDPOINT }}
+        run: npm run eval:compression -- --no-progress-bar -o evals/results/compression-results.json
+        continue-on-error: true
+
+      - name: Upload Evaluation Results
+        uses: actions/upload-artifact@v4
+        with:
+          name: eval-results
+          path: evals/results/*.json
+          retention-days: 30
+
+      - name: Generate Summary
+        run: |
+          echo "## Prompt Evaluation Results" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+
+          for file in evals/results/*.json; do
+            if [ -f "$file" ]; then
+              name=$(basename "$file" .json)
+              echo "### $name" >> $GITHUB_STEP_SUMMARY
+
+              # Extract pass/fail counts using node
+              node -e "
+                const fs = require('fs');
+                const data = JSON.parse(fs.readFileSync('$file', 'utf-8'));
+                const results = data.results || [];
+                const passed = results.filter(r => r.success).length;
+                const failed = results.filter(r => !r.success).length;
+                const total = results.length;
+                const passRate = total > 0 ? ((passed / total) * 100).toFixed(1) : 0;
+                console.log('- Total tests: ' + total);
+                console.log('- Passed: ' + passed);
+                console.log('- Failed: ' + failed);
+                console.log('- Pass rate: ' + passRate + '%');
+              " >> $GITHUB_STEP_SUMMARY 2>/dev/null || echo "- Could not parse results" >> $GITHUB_STEP_SUMMARY
+
+              echo "" >> $GITHUB_STEP_SUMMARY
+            fi
+          done
+
+      - name: Comment on PR
+        if: github.event_name == 'pull_request'
+        continue-on-error: true
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const fs = require('fs');
+            const path = require('path');
+
+            const resultsDir = 'evals/results';
+            if (!fs.existsSync(resultsDir)) {
+              console.log('No results directory found, skipping comment');
+              return;
+            }
+
+            const files = fs.readdirSync(resultsDir).filter(f => f.endsWith('.json'));
+            if (files.length === 0) {
+              console.log('No result files found, skipping comment');
+              return;
+            }
+
+            let summary = '## Prompt Evaluation Results\n\n';
+
+            for (const file of files) {
+              try {
+                const data = JSON.parse(fs.readFileSync(path.join(resultsDir, file), 'utf-8'));
+                const results = data.results || [];
+                const passed = results.filter(r => r.success).length;
+                const total = results.length;
+                const passRate = total > 0 ? ((passed / total) * 100).toFixed(1) : 0;
+                const icon = passRate >= 80 ? '✅' : passRate >= 50 ? '⚠️' : '❌';
+
+                summary += `### ${icon} ${file.replace('.json', '')}\n`;
+                summary += `- Pass rate: **${passRate}%** (${passed}/${total})\n\n`;
+              } catch (e) {
+                summary += `### ❓ ${file}\n- Could not parse results\n\n`;
+              }
+            }
+
+            await github.rest.issues.createComment({
+              issue_number: context.issue.number,
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              body: summary
+            });

.github/workflows/test.yml

@@ -0,0 +1,35 @@
+name: Tests
+
+on:
+  push:
+    branches:
+      - main
+      - master
+  pull_request:
+    branches:
+      - main
+      - master
+
+jobs:
+  test:
+    name: Run Tests
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+          cache: 'npm'
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Run type check
+        run: npm run typecheck
+
+      - name: Run unit tests
+        run: npm test

.gitignore

@@ -1,4 +1,22 @@
+.tts
+.reflection
+.opencode/
 node_modules/
+__pycache__/
 *.log
 .DS_Store
 .env
+
+# Build artifacts
+*.js
+!jest.config.js
+*.js.map
+*.d.ts
+
+# Test artifacts
+fixtures/
+test/mocks/
+
+# Promptfoo eval results
+evals/results/
+evals/evals/