Moxie Docs
Moxie Docs indexes GitHub repos, generates source-cited docs, flags drift on code changes, and provides MCP-ready repo context plus docs-only cleanup PRs.
What is Moxie Docs?
Moxie Docs is a documentation system for GitHub repositories that automatically generates and maintains codebase documentation. It indexes a repo, creates searchable docs for architecture, conventions, and walkthroughs, and rechecks those pages when code changes so outdated material can be flagged and rewritten.
It also exposes repo conventions and verified context to AI tools over MCP, with read-only access scoped to the task. In addition, it can open a weekly docs-only “Friday Cleanup” pull request to collect drift that accumulated during the week.
Key Features
- Continuous repository indexing: Moxie connects to a GitHub repo and keeps an index current as merges happen, so documentation stays tied to the latest code.
- Generated, source-cited docs: It creates architecture pages, conventions, and walkthroughs with citations back to the source code, making the output easier to verify.
- Drift detection on merge: When code changes affect existing docs, Moxie flags the drift and regenerates the relevant pages instead of leaving them stale.
- MCP context for agents: AI tools such as Cursor, Claude Code, and Codex can pull conventions, docs, and verified commands over MCP without recrawling the repo on every prompt.
- Friday Cleanup PRs: Once a week, Moxie can open a docs-only pull request to collect and present documentation updates for review.
- PR workflow checks: It can check pull requests for doc impact and description alignment before merge, while leaving code changes and merge control with the team.
How to Use Moxie Docs
Start by connecting a GitHub repository. Moxie indexes the codebase, generates the first set of documentation, and summarizes conventions that can be reused by humans or AI agents.
From there, teams use it in two main ways: review docs that Moxie generates and updates, and let it check new pull requests for stale documentation or mismatched descriptions. If desired, the MCP integration can be used by coding agents to retrieve repo-specific context in a scoped, read-only way.
Use Cases
- Keeping architecture docs current: A team maintains pages such as webhook behavior or billing flows, with Moxie updating the docs when related code changes land.
- Feeding coding agents repo context: Developers using Cursor, Claude Code, or Codex can retrieve conventions and verified commands without manually browsing the repository.
- Reviewing documentation drift weekly: Teams that want a regular cleanup cadence can review the docs-only Friday Cleanup PR instead of chasing stale pages ad hoc.
- Checking PR impact before merge: Maintainers can see whether a pull request affects documentation or needs a description rewrite before it is merged.
- Onboarding new engineers: New contributors can search generated docs and bookmarks in one workspace instead of reading multiple source files to understand conventions.
FAQ
Does Moxie change my code automatically? No. The source says code changes remain under team control. Documentation updates are delivered as reviewable docs-only pull requests, and description alignment edits the PR description rather than the code.
What happens after I connect a repository? Moxie indexes the repo, generates searchable documentation, summarizes conventions, and starts checking new pull requests for doc impact and description alignment.
Is access to the repository scoped? Yes. The product description says access is read-only by default and scoped per repository, and MCP context is limited to the task rather than the full codebase.
What kinds of documentation does it generate? The page specifically mentions architecture pages, conventions, walkthroughs, and docs for topics such as billing webhooks.
Does it work only with GitHub? The page is written for GitHub repositories, so GitHub is the supported source described here.
Alternatives
- Manual documentation in Markdown or a docs wiki: This is the simplest alternative, but it does not automatically recheck documentation against code or generate source-cited updates.
- General-purpose code search and repository browsing tools: These help people inspect files, but they do not provide a maintained documentation layer or docs-only cleanup PRs.
- Standalone AI coding assistants without repo-aware docs: These can answer questions and edit code, but they still rely on the model rediscovering context unless paired with a system like Moxie.
- Internal docs workflows built around pull requests and review: Teams can manage docs manually through PRs, but they must create and track drift checks themselves rather than having them generated from the codebase.
Alternatives
Theneo
Theneo is an all-in-one developer portal for API docs, guides, changelogs, and private customer docs—built for real-time co-editing.
Ably Chat
Ably Chat is a chat API and SDKs for building custom realtime chat apps, with reactions, presence, and message edit/delete.
Falconer
Falconer is a self-updating knowledge platform for high-speed teams to write, share, and find reliable internal documentation and code context in one place.
AakarDev AI
AakarDev AI is a powerful platform that simplifies the development of AI applications with seamless vector database integration, enabling rapid deployment and scalability.
DeepMotion
DeepMotion is an AI motion capture and body-tracking platform to generate 3D animations from video (and text) in your web browser, via Animate 3D API.
Arduino VENTUNO Q
Arduino VENTUNO Q is an edge AI computer for robotics, combining AI inference hardware and a microcontroller for deterministic control. Arduino App Lab-ready.