licit Documentation
CLI tool for regulatory compliance and code traceability for AI-assisted development teams.
Table of Contents
For users
| Document | Description |
|---|
| Quick Start | Guide to get licit running in 5 minutes |
| CLI Guide | Complete reference of all commands and options |
| Configuration | Configuration guide for .licit.yaml with all fields |
| Connectors | Architect and vigil: what they read, how to configure them, how they feed compliance |
| Examples and recipes | Complete workflows for common use cases |
| Best practices | Recommendations for integrating licit into your workflow |
| FAQ | Frequently asked questions and troubleshooting |
For compliance and auditing
| Document | Description |
|---|
| Compliance | Supported regulatory frameworks: EU AI Act and OWASP Agentic Top 10 |
| Legal framework | Legal context of the EU AI Act, OWASP, NIST, ISO — with official references |
| Report interpretation | How to read and act on compliance and gap analysis reports |
| FRIA Guide | Question-by-question guidance for completing the FRIA (Art. 27) |
| Auditor guide | Compliance verification, evidence, preparation for regulatory audit |
| CI/CD Integration | GitHub Actions, GitLab CI, Jenkins — licit as a compliance gate |
| Enterprise guide | Organizational adoption, maturity model, GRC integration |
For developers
| Document | Description |
|---|
| Architecture | System architecture, modules, phases and design decisions |
| Data models | Enums, dataclasses and Pydantic schemas used internally |
| Provenance | Traceability system: heuristics, git analyzer, store, attestation |
| Changelog | Changelog system: watcher, differ, classifier, renderer |
| Programmatic API | Using licit from Python: imports, classes, examples |
| Security | Threat model, cryptographic signing, data protection |
| Development | Contributor guide: setup, testing, linting, conventions |
| Migration V0 → V1 | Stability contract, planned changes, migration steps |
Reference
| Document | Description |
|---|
| Glossary | Regulatory, technical and domain terms |
Quick Start
# Install
pip install licit-ai-cli
# Initialize in your project
cd your-project/
licit init
# View status
licit status
# Track code provenance
licit trace --stats
# Generate compliance report
licit report
Current version
- v1.0.0 — Stable release. Phases 1-7 completed + exhaustive QA (142 manual tests x 5 projects x 10 edge cases)
- Python 3.12+ required
- 10 CLI commands, all functional (
fria --auto for CI/CD)
- 789 tests, mypy strict, ruff clean
- Connectors: architect (reports, audit log, config) + vigil (SARIF, SBOM)
License
MIT — see LICENSE in the project root.