# Codencer

> Codencer is an open-source orchestration bridge that lets any AI planner (ChatGPT, Claude, Gemini, human) dispatch real work to any AI coding executor (Codex, Claude Code, Antigravity, Qwen, OpenClaw) and produces a durable, auditable run record. It is not a coding agent, not a planner, not an LLM host — it is the bridge. Apache 2.0. Public OSS beta v0.2.0.

## Docs
- [Overview](/docs/overview): What Codencer is and what it isn't.
- [Quickstart](/docs/quickstart): Local daemon + CLI in under 60 seconds.
- [Architecture](/docs/architecture): Planner → bridge → executor, the run lifecycle.
- [State model](/docs/state-model): Runs, steps, attempts, artifacts, validations, gates.
- [Adapters](/docs/adapters): Codex, Claude Code, Antigravity, Qwen, OpenClaw.
- [Self-hosting](/docs/self-hosting): Local-only, relay/runtime, cloud control plane.
- [Security & audit](/docs/security): Trust model, what's in scope, what's deferred.

## Product
- [How it works](/how-it-works): The picture before the prose.
- [Use cases](/use-cases): Three concrete scenarios.
- [For teams](/for-teams): Engineering leaders, governance, audit.

## Optional
- [Sponsor](/sponsor): GitHub Sponsors + community.
- [About](/about): Founder + the three products.
- [Changelog](/changelog): Releases.
- [Blog](/blog): Writing.