# Agenomic

> The electronic genome for production AI agents — behavioral versioning, replay, and governance, with signed, auditable evidence that maps to EU AI Act Articles 9–15.

Agenomic encodes every AI agent into a verifiable genome: a portable, signed record of how an agent reasons, acts, calls tools, and follows policy. Teams version and diff agents by behavior, replay them statistically (comparing output distributions and behavior-contract outcomes rather than asserting bit-identical results), and generate evidence for release governance and EU AI Act compliance. The CLI and SDKs run fully offline with no account; Agenomic Cloud is optional — for sharing signed bundles, review, and replay.

## Documentation
- [Quickstart](http://localhost:3000/docs/quickstart): Version, trace, replay, and prove your first AI agent release: create an Agent Genome, run with signed tracing, verify the Evidence Ledger, generate an Evidence Package, and connect Agenomic over MCP.
- [CLI reference](http://localhost:3000/docs/cli-reference): Core `agenomic` commands for local release, replay, diff, and cloud publishing.
- [Python SDK](http://localhost:3000/docs/python-sdk): Use Python to submit runs, attach metadata, and query release evidence.
- [Cloud guide](http://localhost:3000/docs/cloud-guide): Connect teams, registries, approvals, and AI Act evidence packs.
- [Concepts](http://localhost:3000/docs/concepts): Agents, workflows, runs, bundles, locks, and replay distributions.
- [EU AI Act compliance for AI agents](http://localhost:3000/docs/eu-ai-act-compliance): How Agenomic produces signed, replayable evidence that maps to EU AI Act obligations (Articles 9–15) for production AI agents. Agenomic provides evidence; it does not issue legal certification.
- [EU AI Act technical documentation for AI agents (Article 11 & Annex IV)](http://localhost:3000/docs/ai-act-technical-documentation): Article 11 and Annex IV of the EU AI Act require technical documentation drawn up before a system reaches the market and kept up to date. Here is how an Agenomic genome and bundle capture much of it automatically.
- [Build an audit trail for AI agents (EU AI Act Article 12)](http://localhost:3000/docs/agent-audit-trail): Article 12 of the EU AI Act requires automatic record-keeping over a system’s lifetime. Here is how Agenomic’s signed traces, attestations, and replay give AI agents a tamper-evident audit trail.

## Product
- [Overview](http://localhost:3000/): Behavioral versioning, governance, and EU AI Act evidence for production AI agents.
- [Changelog](http://localhost:3000/changelog): Release notes and shipped capabilities.

## Key facts
- Replay is statistical: it compares output distributions and behavior-contract outcomes across re-runs, not bit-for-bit equality.
- Agenomic produces evidence that maps to EU AI Act Articles 9–15 (risk management, technical documentation, record-keeping); it does not issue a legal certification.
- The CLI and SDKs are fully offline and need no account; Cloud is optional.
- An agent genome captures prompts, models and configuration, tools, permissions, policies, dependencies, memory contracts, execution traces, behavioral fingerprints, and compliance evidence.