A tiny AGENTS.md framework that gives AI agents precise project context through a hierarchy of markdown files.
dox is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo clone agent0ai/doxOpen or clone the template repository.
Move the relevant AGENTS.md content into your project root.
Update project-specific commands, environment rules, and sensitive areas.
Repository setup guidance
Strong trust signals; still review the README and permissions before production use.
Last commit was about 40 days ago.
1402 GitHub stars indicate community interest.
6 open issues signal maintenance load.
MIT license detected.
2 security/trust notes recorded.
Setup difficulty is 1/5.
An MCP server that exposes llms.txt files to IDEs for context-aware development.
Optimize markdown documentation for LLMs and RAG systems, reducing token consumption by 67-95%.
Turn OpenAPI, MCP, Doxygen, godoc, rustdoc, and Markdown into static documentation sites you own.
Local-first MCP server that turns any public GitHub, GitLab, or Bitbucket repo into fresh, version-pinned documentation for AI coding assistants.
DOX is a simple way to help AI coding tools understand your project. You add a special file called AGENTS.md to your project, and the AI agent uses it to keep a tree of documentation files. Before making changes, the agent reads the relevant docs, so it knows the rules and doesn't mess up your code. It's just markdown, no installation needed.
DOX is a lightweight, self-documenting framework built around AGENTS.md files. It provides a structured approach for AI agents to maintain and use project documentation. The core idea is a hierarchy: a root AGENTS.md contains project-wide instructions and an index, while child AGENTS.md files hold local guidelines for specific areas. Before any edit, the agent traverses the docs tree from root to the relevant area, ensuring it follows local rules. After meaningful changes, it updates the affected files, keeping documentation current. This reduces guesswork and drift in AI-assisted development. DOX requires no installation, dependencies, or runtime—it's just a markdown instruction. It works with any AI agent that supports AGENTS.md, such as Codex, Claude Code, and OpenCode. To use it, copy the provided AGENTS.md into your project root; the agent will then start building the DOX tree. For existing projects, you can prompt the agent to 'Initialize DOX tree for this project now.' DOX is created by Agent Zero, an open-source agentic AI framework.
1,402
Stars
188
Forks
6
Issues
MIT
License
A tiny AGENTS.md framework that gives AI agents precise project context through a hierarchy of markdown files.
Developers using AI coding assistants like Codex or Claude Code., Tech leads who want to enforce coding standards via documentation., Open-source maintainers seeking to guide AI contributors.
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
No security risks identified; it's just markdown instructions. Ensure AGENTS.md content is trusted; malicious instructions could mislead AI agents.
Similar or complementary options to evaluate include mcpdoc, llm-docs-builder, sourcey.
Aug 20, 2026