agentleFS
SwastikPandey1024SwastikPandey1024/agentic-engineering-os/.agents/skills/release-engineering/SKILL.md

release-engineering

Skill1 starsChanged 3 months agoMITView on GitHub

Semantic versioning (SemVer), release lifecycle gates, automated changelog generation, rollback runbooks, and verified production deployments.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 10
  2. Distinctive:

    Never cut a release without passing 100% of automated tests, security scans, and clean build checks.

    line 16
  3. Very specific:

    Current Version: Check pyproject.toml, package.json, or Git tags (git describe --tags --abbrev=0).

    line 22
  4. Distinctive:

    CI Pipeline Status: Verify that the latest commit on main passed all automated CI workflows.

    line 24
  5. Distinctive:

    NEVER tag a release on a dirty working tree: Ensure git status is clean.

    line 111
  6. Very specific:

    NEVER manually edit git tags after pushing: If a release contains a bug, release a subsequent patch version (v1.0.1), never overwrite …

    line 112

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Release Engineering Skill

1. When Should I Use This?

Use this skill when:

  • Cutting a new release or version tag (v1.0.0, v1.1.0, v1.0.1) for a library, microservice, or web application.
  • Updating CHANGELOG.md following the Keep a Changelog standard.
  • Executing the pre-flight release gate (Tests → Lint → Security → Build → Versioning).
  • Planning or executing a production rollback following a failed deployment.

Never cut a release without passing 100% of automated tests, security scans, and clean build checks.


2. What Should I Inspect First?

  1. Current Version: Check pyproject.toml, package.json, or Git tags (git describe --tags --abbrev=0).
  2. Unreleased Changes: Inspect git log <last-tag>..HEAD to classify changes as Major (breaking), Minor (new feature), or Patch (bug fix).
  3. CI Pipeline Status: Verify that the latest commit on main passed all automated CI workflows.

3. What Workflow Should I Follow?

1. CHANGE: All features & fixes merged into main
      ↓
2. TEST & LINT: Run full test suite & linters (pytest, ruff, npm test)
      ↓
3. SECURITY: Run vulnerability & secret scans (pip-audit, gitleaks)
      ↓
4. BUILD: Produce & validate build artifacts (Docker image, dist/ wheels)
      ↓
5. VERSION: Bump SemVer in pyproject.toml / package.json
      ↓
6. CHANGELOG: Document Added, Changed, Deprecated, Removed, Fixed
      ↓
7. RELEASE: Create signed Git tag (git tag -a vX.Y.Z -m "Release vX.Y.Z")
      ↓
8. DEPLOY: Trigger deployment to staging/production
      ↓
9. SMOKE TEST: Probe live health endpoints and critical user journeys
      ↓
10. OBSERVE: Monitor error logs and metrics for 15 minutes post-deploy
      ↓
(ROLLBACK if health checks fail or error rates spike > 1%)

Semantic Versioning (SemVer) Rules

Given version MAJOR.MINOR.PATCH (e.g. 1.2.3):

1. MAJOR (e.g. 2.0.0): Incompatible API changes, breaking database migrations.
2. MINOR (e.g. 1.3.0): Backward-compatible new functionality or endpoints.
3. PATCH (e.g. 1.2.4): Backward-compatible bug fixes or security patches.

Standard Changelog Entry (CHANGELOG.md)

## [1.1.0] - 2026-08-14

### Added
- Multi-modal OCR bounding box preservation in `OCRService`.
- FAISS vector store local persistence with automated reload on startup.

### Changed
- Migrated default package management toolchain to `uv`.

### Fixed
- Resolved duplicate timestamp aggregation during Daylight Saving Time fallback.

### Security
- Upgraded `cryptography` dependency to resolve CVE-2026-XXXX.

Production Rollback Procedure

# 1. Identify previous stable release tag
git tag -l "v*" --sort=-v:refname | head -n 5

# 2. Revert deployment trigger
git checkout v1.0.0

# 3. Redeploy previous container image or redeploy commit on Render/Vercel
# 4. Run post-rollback health verification
curl -i https://app-backend.onrender.com/api/v1/health

4. What Decisions Should I Make?

Release DecisionStandard Heuristic
Release Tag FormatUse prefixed semantic versioning (e.g. v1.2.0).
Breaking Database MigrationsImplement two-phase migrations (Expand-Contract pattern) so old code can run alongside new schema before dropping columns.
Release Gate FailureIf any single test, lint check, or security scan fails, ABORT the release immediately.

5. What Should I Avoid?

  • NEVER tag a release on a dirty working tree: Ensure git status is clean.
  • NEVER manually edit git tags after pushing: If a release contains a bug, release a subsequent patch version (v1.0.1), never overwrite v1.0.0.
  • NEVER declare deployment success without executing post-deployment smoke tests.

6. How Should I Verify Success?

# 1. Verify version numbers match across all manifests
python -c "
import tomllib
pyproj = tomllib.load(open('pyproject.toml', 'rb'))['project']['version']
print(f'pyproject.toml version: {pyproj}')
"

# 2. Verify git tag creation
git describe --tags --exact-match

# 3. Verify live endpoint response post-deployment (Cross-Platform)
python -c "import urllib.request, json; res = urllib.request.urlopen('https://app-backend.onrender.com/api/v1/health'); data = json.load(res); assert data.get('status') == 'ok', f'Unexpected status: {data}'; print('Post-release health probe: OK')"

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.