ai-engineering 0.8.4

AI governance framework for secure software delivery

{ai} engineering turns AI-assisted delivery into a governed local workflow.

53 skills · 9 agents · 6 surfaces · 1 governed flow

{ai} engineering installs a deterministic governance layer into any repository: specs, decisions, skills, agents, runbooks, hooks, and audit trails as versioned local files. No hosted control plane. No provider lock-in. Every IDE follows the same rules.

Install

1.1 — Prerequisites

Python 3.11+ and Git.

1.2 — Install ai-engineering

1. If you use uv, install the published PyPI release as a managed tool:

uv tool install --force ai-engineering
uv tool update-shell
exec "$SHELL" -l

2. If you do not use uv, install with pipx:

pipx install ai-engineering

3. If you do not use uv or pipx, install with pip:

python -m pip install --user ai-engineering

If ai-eng is not available after the pip install, add Python's user scripts directory to your PATH, then reopen your shell.

4. Verify the CLI:

ai-eng version

5. Install governance into a repository:

cd your-project
ai-eng install .
ai-eng doctor

[PASS] doctor confirms hooks, mirrors, manifest defaults, and required tools.

To update later from PyPI:

1. Upgrade the installed CLI with the command that matches your install method:

uv tool upgrade ai-engineering
# or
pipx upgrade ai-engineering
# or
python -m pip install --user --upgrade ai-engineering

These commands fetch the latest published ai-engineering release from PyPI.

2. Verify that the new version is available:

ai-eng version

3. In each governed project, refresh the installed framework files:

cd your-project
ai-eng update
ai-eng doctor

Governed Flow

The canonical chain is:

/ai-brainstorm → /ai-plan → /ai-build → /ai-pr

Use it when work changes product behavior, framework behavior, security posture, public docs, or release state. /ai-commit remains available for WIP checkpoints; it is not part of the canonical delivery chain.

Supported Surfaces

One canonical payload is mirrored into all enabled surfaces:

Surface	Entry point
Claude Code	CLAUDE.md
GitHub Copilot	.github/copilot-instructions.md
OpenAI Codex	AGENTS.md
Antigravity	AGENTS.md + .agents/ skills and agents
OpenCode	.opencode/ skills and commands
Cursor	.cursor/ skills

The ruleset lives in AGENTS.md. Project identity and hard prohibitions live in CONSTITUTION.md. Release history and breakage notes live in CHANGELOG.md.

Why Governance Matters

• Spec-driven work keeps LLM output tied to approved scope.
• Deterministic gates catch secrets, broken mirrors, missing docs, and policy drift.
• The local NDJSON audit chain records what happened without sending telemetry by default.
• Skills and agents are file-backed, reviewable, and synchronized across IDEs.

Standing on the shoulders of...

ai-engineering builds on ideas, patterns, and principles from these projects:

Project	What we learned
Superpowers	Brainstorm hard-gate, TDD-for-skills patterns
review-code	Handler-as-workflow architecture, parallel specialist agents, finding-validator
dotfiles/ai	Agent matrix, SDLC coverage patterns
autoresearch	Radical simplicity as a design principle
Emil Kowalski	Motion principles, spring physics, easing strategy
SpecKit	Spec-driven workflow inspiration
GSD	Autonomous execution patterns
Anthropic Skills	Frontend-design, canvas, skill-creator — absorbed and extended

Contributing

Contributions are welcome. See CONTRIBUTING.md for development setup, code style, testing, and the pull request process.

Code of conduct

This project follows the Contributor Covenant Code of Conduct. See CODE_OF_CONDUCT.md.

License

MIT. See LICENSE.