PR Review Workflow — Agent Summarizes the Diff Before I Read a Single Line

# Fetch PR details
pr_data=$(gh pr view "$PR_NUMBER" --repo "$REPO" \
  --json title,body,files,additions,deletions,baseRefName,headRefName,\
commits,statusCheckRollup,reviews,linkedIssues)

# Get the full diff
gh pr diff "$PR_NUMBER" --repo "$REPO" > /tmp/pr_diff.patch

📝 PR #156: "Add rate limiting to API endpoints"

Summary of changes:
- New middleware: RateLimitMiddleware in api/middleware/rate_limit.py
- Configuration: Rate limits defined per-endpoint in settings.py
- Redis integration: Uses Redis for distributed rate counting
- 3 endpoint decorators updated to include rate limit headers
- New management command for resetting rate limit counters

RISK_FACTORS = {
    "high": {
        "patterns": ["middleware", "auth", "security", "payment", "migration"],
        "file_paths": ["settings.py", "urls.py", "wsgi.py", "manage.py"]
    },
    "medium": {
        "patterns": ["model", "serializer", "view", "permission"],
        "thresholds": {"files_changed": 10, "additions": 300}
    }
}

def assess_risk(pr_data: dict, diff: str) -> str:
    risk_flags = []
    for file in pr_data["files"]:
        path = file["path"].lower()
        if any(p in path for p in RISK_FACTORS["high"]["patterns"]):
            risk_flags.append(f"⚠️ HIGH: Modifies {file['path']}")

    if pr_data["additions"] > 300 and pr_data["deletions"] < 20:
        risk_flags.append("⚠️ MEDIUM: Large addition with minimal cleanup")

    return risk_flags

🎯 Risk Assessment: MEDIUM-HIGH
  ⚠️ HIGH: Modifies api/middleware/ (request pipeline)
  ⚠️ MEDIUM: Touches settings.py (configuration)
  ✅ LOW: Test files have matching additions

  Flags:
  - This PR adds middleware that runs on EVERY request.
    Performance impact should be benchmarked.
  - Redis dependency added — ensure Redis is in all environments.

# Check if new code has corresponding tests
new_files=$(echo "$pr_data" | jq -r '.files[].path' | grep -v test)
test_files=$(echo "$pr_data" | jq -r '.files[].path' | grep test)

# Count new functions/classes vs new test functions
new_functions=$(grep -c "def \|class " /tmp/pr_diff.patch | grep "^+")
new_tests=$(grep -c "def test_\|class Test" /tmp/pr_diff.patch | grep "^+")

🧪 Test Coverage:
  New code: 4 functions, 1 class added
  New tests: 6 test functions added
  Coverage ratio: 1.5 tests per function ✅

  Missing coverage:
  - No test for rate_limit_exceeded response format
  - No test for Redis connection failure fallback

🔄 CI Status: ✅ All 4 checks passing
  - lint: ✅ (12s)
  - test-unit: ✅ (1m 34s)
  - test-integration: ✅ (2m 12s)
  - build: ✅ (3m 01s)

📋 Merge Status:
  - Conflicts: None
  - Branch freshness: 2 commits behind main (rebase recommended)
  - Linked issues: #148 (rate limiting request)

🏁 Recommendation: REQUEST CHANGES

Reasoning:
1. Core logic looks correct and well-structured
2. Missing tests for error paths (Redis down, malformed headers)
3. No performance benchmark for middleware on hot path
4. settings.py change needs env variable documentation

Suggested action: Ask for 2 additional tests + performance note
in PR description. Approve after.