Fix Ci

# Fix CI Failures (Autonomous)

Automatically fetch failed GitHub Actions workflow logs, analyze issues, fix them, validate, commit, push, and repeat until CI is green. **No user input required** — runs fully autonomously.

## Context

- **Repository**: `sarthchawla/InZone`
- **Current branch**: `{{ git.currentBranch }}`
- **CI Platform**: GitHub Actions

## Project Structure

InZone/ ├── apps/ │ ├── web/ # React frontend (Vite) │ └── api/ # Node.js backend (Express) ├── packages/ │ └── shared/ # Shared types/utilities └── .github/workflows/ ├── ci-backend.yml # Backend lint + build + unit tests ├── ci-frontend.yml # Frontend lint + build + unit tests ├── bdd-backend.yml # Cucumber BDD API tests ├── bdd-frontend.yml # Playwright-BDD E2E tests ├── architecture-tests.yml └── worktree-scripts.yml

## Autonomous Loop

**Repeat the following loop until all workflows pass or you've completed 5 iterations:**

### Step 1: Get Latest Workflow Runs

```bash
gh run list --branch <current-branch> --limit 10

Find the most recent set of workflow runs (same timestamp group). If all are success, report success and stop.

Step 2: Get Failed Job Logs

For each failed workflow run:

gh run view <run-id> --log-failed

If no failed jobs, all workflows passed — report success and stop.

Step 3: Analyze Failures

For each failed job, identify the issue type:

Error Detection Patterns:

Pattern	Type
error TS\d+:	TypeScript compiler error
Type '.*' is not assignable	Type mismatch
Cannot find module	Missing import
Property '.*' does not exist	Missing property
FAIL, AssertionError, Expected:, Received:	Vitest test failure
FAILED, Timeout, expect.toBeVisible, locator.click	Playwright-BDD failure
error with rule name	ESLint error
PrismaClientKnownRequestError, P\d{4}	Prisma/Database error
Connection refused	Database connection issue

Step 4: Present Analysis Summary

Show a brief summary table:

## CI Failure Analysis (Iteration N)

| Workflow | Job | Failure Type | Files Affected |
|----------|-----|--------------|----------------|
| ... | ... | ... | ... |

Step 5: Apply Fixes

Do not ask the user — automatically fix all issues:

1. Read each affected file
2. Apply the fix using the Edit tool
3. Briefly describe what was changed

Fix Priority:

• TypeScript/build errors first (they block everything)
• ESLint errors second
• Test assertion mismatches third (update test expectations or fix source code)
• Workflow/config issues fourth

Step 6: Validate Fixes Locally

Run appropriate validation based on failure type:

# TypeScript/Build errors
pnpm --filter <package> build

# Lint errors
pnpm --filter <package> lint

# Frontend Unit Test failures
pnpm --filter web test run -- <test-file>

# Backend Unit Test failures
pnpm --filter api test run -- <test-file>

# Type checking
pnpm --filter <package> typecheck

If local validation fails, fix and re-validate before proceeding.

Step 7: Commit and Push

Automatically stage, commit, and push:

1. Stage only the changed files: git add <specific-files>
2. Commit with descriptive message:

   fix: resolve CI failures from workflow run #<id>
   
   - <brief description of fix 1>
   - <brief description of fix 2>
   
   Co-Authored-By: Claude <noreply@anthropic.com>
   

3. Push: git push

Step 8: Watch CI

Poll workflow runs until all complete:

# Poll every 20 seconds until no in_progress runs remain
gh run list --branch <branch> --limit 10 --json status,name,conclusion,databaseId

Step 9: Check Results

• If all workflows pass: Report success and stop
• If any workflow fails: Go back to Step 2 (next iteration)
• If 5 iterations reached: Report remaining failures and stop

Guidelines

• Always read files before editing
• Run local validation after fixes before pushing
• Never suppress lint errors — fix root cause
• For test failures, determine if test or code needs fixing
• Keep commit messages concise but descriptive
• Each iteration should make a single focused commit
• Do NOT ask the user any questions — make reasonable decisions autonomously

Useful Commands

# List recent workflow runs
gh run list --branch $(git branch --show-current) --limit 10

# View specific run
gh run view <run-id>

# Get failed logs
gh run view <run-id> --log-failed

# Rerun failed jobs
gh run rerun <run-id> --failed

# Watch a running workflow
gh run watch <run-id>