cimonitor

name	cimonitor
description	Monitor CI pipeline status, diagnose failures, and drive the badge back to green

Monitor CI/CD pipeline status. Checks both the current branch and the CI badge workflow on main.

1. Check the CI badge workflow (always)

The README badge tracks the CI workflow on main. Always check this first:

gh run list --workflow ci.yml --branch main --limit 1 --json status,conclusion,databaseId,url

Report whether the badge is green or red. If red, this takes priority — investigate and fix before other work.

2. Find runs for the current branch

gh run list --branch $(git branch --show-current) --limit 5 --json status,conclusion,name,url,databaseId,event,workflowName

3. Poll in-progress runs

For each in_progress or queued run, poll every 15 seconds up to 20 attempts:

gh run view <run-id> --json status,conclusion,name

Show a brief status update each poll.

4. Per-job breakdown

For each completed run, show individual job results:

gh run view <run-id> --json jobs --jq '.jobs[] | {name, conclusion, startedAt, completedAt}'

Present as a table:

Job	Status	Duration
Python Tests	success	2m 15s
Security Tests	success	1m 30s
...	...	...

The CI workflow has these jobs (all must pass for a green badge):

• python-tests — Unit tests with coverage (~8000 tests)
• security-tests — JWT, ASVS, sanitization, bandit
• lint — ruff + bandit + DSL validation + AsyncAPI
• type-check — mypy on core + backend
• integration — Integration tests (depends on tests + lint + security)
• e2e-smoke — Example project validation + serve tests
• postgres-tests — Full test suite against PostgreSQL
• e2e-runtime — CRUD + DSL tests against PostgreSQL
• homebrew-validation — Formula syntax + version consistency

5. Diagnose and fix failures

For each failed job:

1. Fetch logs: gh run view <run-id> --log-failed | tail -100
2. Categorize the failure:
   • Type error (mypy) → fix locally with mypy src/dazzle (matches /ship, /check, and CI)
   • Lint failure (ruff) → fix locally with ruff check src/ tests/ --fix && ruff format src/ tests/
   • Test failure → identify the test, run locally with pytest <test_file> -x -v
   • Security failure (bandit) → check the flagged code
   • Flaky/infra (timeout, network) → note as transient, suggest re-run
3. Fix ALL errors — including pre-existing ones. The goal is a green badge, not just validating your own changes. If the CI was already red before your push, fix those errors too. Pre-existing mypy errors, lint warnings, or test failures should be fixed in a separate commit with a message like fix: resolve pre-existing mypy errors in <file>.
4. Commit, push, and re-poll until the badge is green.

6. Report summary

## CI Status

**Badge (main):** ✅ green | ❌ red (failing since <date>)
**Current branch:** <branch> — <status>

### Job Results (run #<id>)
| Job | Result |
|-----|--------|
| ... | ...    |

### Action Required
- <what needs fixing, if anything>