Skip to content

High-performance browser automation bridge and multi-instance orchestrator with advanced stealth injection and real-time dashboard.

revert: remove AI code review feature, keep only draft PR skip

Datafe5b0c87338c
diff --git a/.github/scripts/ai-review.js b/.github/scripts/ai-review.js
deleted file mode 100644
index 65533e91..00000000
--- a/.github/scripts/ai-review.js
+++ /dev/null
@@ -1,107 +0,0 @@
-#!/usr/bin/env node
-/**
- * AI Code Review Script
- * Sends PR diff to Claude for review and posts results as a comment
- */
-
-const https = require('https');
-const fs = require('fs');
-const { execSync } = require('child_process');
-
-const ANTHROPIC_API_KEY = process.env.ANTHROPIC_API_KEY;
-const GH_TOKEN = process.env.GH_TOKEN;
-const PR_NUMBER = process.env.PR_NUMBER;
-
-if (!ANTHROPIC_API_KEY || !GH_TOKEN || !PR_NUMBER) {
- console.error('Missing required environment variables');
- process.exit(1);
-}
-
-// Read PR diff
-const diff = fs.readFileSync('pr.diff', 'utf8');
-
-// Prepare Claude API request
-const payload = JSON.stringify({
- model: 'claude-sonnet-4-5',
- max_tokens: 4096,
- messages: [{
- role: 'user',
- content: `You are a code reviewer for the pinchtab project. Review this PR diff and provide concise, actionable feedback.
-
-Focus on:
-1. Bugs or logic errors
-2. Security issues (especially command injection, path traversal)
-3. Performance concerns
-4. Code clarity and maintainability
-5. Violations of project conventions (no redundant comments, proper error handling)
-
-Be specific about line numbers. If the code looks good, say so briefly. Keep response under 2000 characters.
-
-PR Diff:
-
-${diff}`
- }]
-});
-
-// Call Claude API
-const options = {
- hostname: 'api.anthropic.com',
- path: '/v1/messages',
- method: 'POST',
- headers: {
- 'x-api-key': ANTHROPIC_API_KEY,
- 'anthropic-version': '2023-06-01',
- 'content-type': 'application/json',
- 'Content-Length': Buffer.byteLength(payload)
- }
-};
-
-console.log('Calling Claude API for code review...');
-
-const req = https.request(options, (res) => {
- let data = '';
-
- res.on('data', (chunk) => {
- data += chunk;
- });
-
- res.on('end', () => {
- try {
- const response = JSON.parse(data);
-
- if (response.error) {
- console.error('Claude API error:', response.error);
- process.exit(1);
- }
-
- const reviewText = response.content[0].text;
- console.log('Review received from Claude');
-
- // Post comment to PR
- const comment = `## πŸ€– AI Code Review (Claude)
-
-${reviewText}
-
----
-*Automated review using Claude Sonnet 4.5*`;
-
- execSync(`gh pr comment ${PR_NUMBER} --body ${JSON.stringify(comment)}`, {
- env: { ...process.env, GH_TOKEN }
- });
-
- console.log('Review posted to PR');
- } catch (err) {
- console.error('Error processing response:', err);
- console.error('Response:', data);
- process.exit(1);
- }
- });
-});
-
-req.on('error', (err) => {
- console.error('Request error:', err);
- process.exit(1);
-});
-
-req.write(payload);
-req.end();
diff --git a/.github/workflows/ai-code-review.yml b/.github/workflows/ai-code-review.yml
deleted file mode 100644
index 55c01efd..00000000
--- a/.github/workflows/ai-code-review.yml
+++ /dev/null
@@ -1,51 +0,0 @@
-name: AI Code Review
-
-on:
- pull_request:
- types: [opened, synchronize, reopened, ready_for_review]
-
-permissions:
- contents: read
- pull-requests: write
-
-jobs:
- review:
- name: Claude Code Review
- runs-on: ubuntu-latest
- if: github.event.pull_request.draft == false
- steps:
- - uses: actions/checkout@v4
- with:
- fetch-depth: 0
-
- - name: Get PR diff
- id: diff
- run: |
- # Get the diff for this PR
- gh pr diff ${{ github.event.pull_request.number }} > pr.diff
-
- # Check if diff is too large (Claude has token limits)
- DIFF_SIZE=$(wc -c < pr.diff)
- if [ "$DIFF_SIZE" -gt 100000 ]; then
- echo "skip=true" >> $GITHUB_OUTPUT
- echo "Diff too large ($DIFF_SIZE bytes), skipping AI review"
- exit 0
- fi
-
- echo "skip=false" >> $GITHUB_OUTPUT
- env:
- GH_TOKEN: ${{ github.token }}
-
- - name: Setup Node.js
- if: steps.diff.outputs.skip == 'false'
- uses: actions/setup-node@v4
- with:
- node-version: '22'
-
- - name: Review with Claude
- if: steps.diff.outputs.skip == 'false'
- run: node .github/scripts/ai-review.js
- env:
- ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
- GH_TOKEN: ${{ github.token }}
- PR_NUMBER: ${{ github.event.pull_request.number }}
diff --git a/docs/guides/ai-code-review.md b/docs/guides/ai-code-review.md
deleted file mode 100644
index fd51c5b1..00000000
--- a/docs/guides/ai-code-review.md
+++ /dev/null
@@ -1,176 +0,0 @@
-# AI Code Review Setup
-
-Automated code review using Claude after CI checks pass.
-
-## How It Works
-
-1. PR is opened/updated and marked "Ready for review"
-2. CI checks run (tests, lint, etc.)
-3. **After CI completes**, Claude reviews the PR diff
-4. Claude posts review as a PR comment
-
-## Setup
-
-### 1. Add Anthropic API Key
-
-Get an API key from https://console.anthropic.com/
-
-Add to repo secrets:
-```bash
-# Via GitHub CLI
-gh secret set ANTHROPIC_API_KEY --body "sk-ant-..."
-
-# Or via web UI:
-# Settings β†’ Secrets and variables β†’ Actions β†’ New repository secret
-# Name: ANTHROPIC_API_KEY
-# Value: sk-ant-...
-```
-
-### 2. Enable the Workflow
-
-The workflow is already configured in `.github/workflows/ai-code-review.yml`.
-
-It runs automatically on:
-- PR opened
-- PR synchronized (new commits)
-- PR reopened
-- PR marked "Ready for review"
-
-**Skips draft PRs** β€” no review until you mark as ready.
-
-## Cost Estimation
-
-**Per PR review:**
-- Input: ~5K-50K tokens (depending on diff size)
-- Output: ~1K tokens (review text)
-- Cost: ~$0.05-$0.30 per review (Sonnet 4.5 pricing)
-
-**Monthly estimate:**
-- 50 PRs/month Γ— $0.15 average = ~$7.50/month
-
-**Large diffs (>100KB) are automatically skipped** to avoid excessive costs.
-
-## Customization
-
-### Change Model
-
-Edit `.github/scripts/ai-review.js`:
-```javascript
-model: 'claude-sonnet-4-5', // Fast, cheaper
-// OR
-model: 'claude-opus-4', // More thorough, expensive
-```
-
-### Change Review Focus
-
-Edit the prompt in `.github/scripts/ai-review.js`:
-```javascript
-content: `Review this PR focusing on:
-1. Security issues
-2. Performance
-3. Maintainability
-...`
-```
-
-### Skip Certain Files
-
-Edit `.github/workflows/ai-code-review.yml`:
-```yaml
-- name: Get PR diff
- run: |
- # Exclude generated files, vendor, etc.
- gh pr diff ${{ github.event.pull_request.number }} \
- -- ':!vendor/**' ':!*.gen.go' ':!dashboard/dist/**' > pr.diff
-```
-
-## Disabling
-
-### Temporarily (per PR)
-
-Add `[skip ai-review]` to PR title or description.
-
-Edit workflow to check for this:
-```yaml
-if: |
- github.event.pull_request.draft == false &&
- !contains(github.event.pull_request.title, '[skip ai-review]')
-```
-
-### Permanently
-
-Delete or rename `.github/workflows/ai-code-review.yml`
-
-## Example Output
-
-```markdown
-## πŸ€– AI Code Review (Claude)
-
-**Security:** Line 327 - exec.Command with user-controlled input. The #nosec
-directive is appropriate here since chromeBinary comes from config/env vars,
-but consider validating the path exists before execution.
-
-**Performance:** Consider caching the Chrome binary lookup result in
-findChromeBinary() to avoid repeated filesystem checks.
-
-**Code Quality:** Well-documented! The goroutine+timer pattern for startup
-timeout is elegant and avoids the context cancellation issue.
-
-Overall this is solid work. The fallback strategy for Chrome 145 is robust
-and backwards compatible.
-
----
-*Automated review using Claude Sonnet 4.5*
-```
-
-## Troubleshooting
-
-### "ANTHROPIC_API_KEY not set"
-
-Secret not configured or workflow doesn't have access.
-- Check: Settings β†’ Secrets β†’ ANTHROPIC_API_KEY exists
-- Verify workflow has `permissions: pull-requests: write`
-
-### "Diff too large, skipping AI review"
-
-PR changes >100KB. Options:
-1. Split into smaller PRs
-2. Increase limit in workflow (line 26)
-3. Review manually
-
-### "Claude API error: rate_limit_error"
-
-API rate limit hit. Options:
-1. Wait a few minutes and re-run workflow
-2. Upgrade Anthropic plan
-3. Reduce review frequency (skip draft PRs, etc.)
-
-## Alternative: OpenAI
-
-Don't have Anthropic API? Use OpenAI instead:
-
-Edit `.github/scripts/ai-review.js`:
-```javascript
-// Change API endpoint
-hostname: 'api.openai.com',
-path: '/v1/chat/completions',
-
-// Change headers
-headers: {
- 'Authorization': `Bearer ${OPENAI_API_KEY}`,
- 'Content-Type': 'application/json'
-}
-
-// Change payload
-const payload = JSON.stringify({
- model: 'gpt-4-turbo',
- messages: [{ role: 'user', content: `...` }]
-});
-```
-
-Update secret name in workflow: `OPENAI_API_KEY`
-
-## Further Reading
-
-- [Anthropic API Docs](https://docs.anthropic.com)
-- [GitHub Actions Docs](https://docs.github.com/actions)
-- [Definition of Done](../../DEFINITION_OF_DONE.md) β€” What Claude checks for
diff --git a/docs/index.json b/docs/index.json
index 46d456a0..5c02d728 100644
--- a/docs/index.json
+++ b/docs/index.json
@@ -28,8 +28,7 @@
"guides/agent-optimization.md",
"guides/memory-monitoring.md",
"guides/data-storage.md",
- "guides/raspberry-pi.md",
- "guides/ai-code-review.md"
+ "guides/raspberry-pi.md"
],
"proposals": [
"proposals/default-instance.md",