# Docs

> Docs is an open-source agent for coding. Self-programming AI agent that writes and documents its own code, aimed at developers. It is available on CLI and self-hosted.

- **Kind:** Agent
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted
- **Website:** https://automata.readthedocs.io/en/latest/
- **Categories:** [coding](https://ai-powered.com/categories/coding-agents)
- **Listing updated:** 2026-11-18

## About

The site describes Automata as an autonomous system that uses large language models such as GPT-4 together with a vector database to document, search and write code. Its stated long-term goal is general AI, though the documentation presents that as an aspiration rather than a current capability.

The project's documentation covers several pieces: code embeddings generated from its own codebase, SCIP indices used for search, an execution layer for running coding tasks through downstream tooling, and an agent that can carry out trivial and non-trivial instructions. Setup guides cover a local environment, environment variables, pre-commit hooks and Git LFS.

The docs say the module reference is generated automatically and is still a work in progress. The project is aimed at developers who want to experiment with an LLM-driven coding agent that operates on its own source.

## Features

- **Code embeddings:** Generates embeddings from the project's codebase so the agent can search and reason over its own code.
- **SCIP-based code search:** Builds SCIP indices that power the Automata search over symbols and references.
- **Code and doc generation:** Produces code and documentation through guided generation workflows.
- **Configurable agent runs:** Runs an agent that handles both simple and multi-step coding instructions.
- **Task execution layer:** Executes coding tasks through downstream tooling, with a task registry and environment.

## Good for

- Experimenting with an LLM agent that edits its own codebase
- Generating documentation for a Python project from its source
- Building semantic code search over a repository using embeddings
- Studying how an autonomous coding agent is structured

## FAQ

### Which language models does Automata use?

The site says it uses large language models such as GPT-4, and the module list includes an OpenAI-based agent.

### What does Automata use for storage and search?

It uses a vector database for embeddings and SCIP indices for code search.

### Is Docs free?

Yes. Docs is open source, so you can use and self-host it for free.

### What are the best alternatives to Docs?

The closest alternatives to Docs are clippinator, gpt-pilot, devika, CoderBaba and ANUS.

## Where this comes from

- Description: written from the project's website.

## Alternatives

- [clippinator](https://ai-powered.com/agents/clippinator): clippinator is an open-source agent for coding.
- [gpt-pilot](https://ai-powered.com/agents/gpt-pilot): gpt-pilot is an open-source agent for coding.
- [devika](https://ai-powered.com/agents/devika): devika is an open-source agent for coding.
- [CoderBaba](https://ai-powered.com/agents/coderbaba): CoderBaba is a free open-source agent for coding.
- [ANUS](https://ai-powered.com/agents/anus): ANUS is a free open-source agent for coding.
- [Agav](https://ai-powered.com/agents/agav): Agav is an open-source agent for coding.
- [DeepCode](https://ai-powered.com/agents/deepcode): DeepCode is an open-source agent for coding.
- [letta](https://ai-powered.com/agents/letta): letta is an open-source agent framework for knowledge and memory.

---
Source: https://ai-powered.com/agents/docs-12
