Moxie Docs

Product Introduction

1. Definition: Moxie Docs is a living documentation and AI context management platform specifically designed for software development teams using GitHub. It is a GitHub App that functions as a documentation automation tool and a context provider for AI coding agents via the Model Context Protocol (MCP).
2. Core Value Proposition: Moxie Docs exists to eliminate stale documentation and improve code quality by creating a single, automatically updated source of truth. Its primary value is ensuring that both human developers and AI agents work from the same accurate, cited, and context-aware information, leading to fewer errors, faster onboarding, and cleaner pull requests.

Main Features

1. MCP Context for AI Agents: This feature delivers scoped, read-only repository context—including conventions, code architecture, and documentation gaps—directly to AI agents like Cursor, Claude Code, Codex, and GitHub Copilot via the Model Context Protocol.
   • How it works: After indexing a repository, Moxie Docs exposes structured data through MCP endpoints (e.g., get_conventions(), get_doc_gaps()). Agents query these endpoints during a coding session, receiving lightweight, cited context instead of performing full, token-expensive codebase re-crawls. This prevents context pollution and ensures agents adhere to project-specific standards from the first prompt.
2. Searchable Docs Workspace: A centralized, searchable knowledge hub that aggregates generated architecture pages, convention summaries, and existing Markdown files with persistent highlighting and bookmarks.
   • How it works: Upon indexing, Moxie Docs scans source code, tests, existing docs, and commit history to generate a living index. This content is presented in a clean interface where every answer is cited back to its source file. New engineers can use this workspace for onboarding, drastically reducing "scavenger hunts" for information and repetitive Slack questions.
3. PR Automation & Doc-Impact Checks: An automated system that ensures pull request descriptions align with team templates and scans for documentation that should be updated based on the code changes.
   • How it works: On every new PR, Moxie Docs runs a description alignment check against a pre-configured template. It also performs a doc-impact scan, flagging stale or missing documentation related to the modified files. Furthermore, a weekly "Friday Cleanup" automatically opens small, reviewable, docs-only pull requests to address accumulated documentation debt, preventing rot.

Problems Solved

1. Pain Point: Documentation rot and context fragmentation, where project knowledge is scattered, outdated, or inaccessible to both developers and AI tools, leading to repeated mistakes, onboarding delays, and inefficient AI-assisted coding.
2. Target Audience: Software Development Teams, Technical Leads, DevOps Engineers, and Full-Stack Developers who use GitHub, maintain complex codebases, and are increasingly integrating AI coding assistants into their workflow. It is especially valuable for teams with multiple services or repositories.
3. Use Cases: Accelerating new engineer onboarding with a single source of truth. Maintaining consistent code quality by giving AI agents project-specific conventions. Automating documentation maintenance to prevent outdated docs. Improving pull request quality and review efficiency through automated alignment and impact checks. Reducing token costs and errors when using AI coding agents.

Unique Advantages

1. Differentiation: Unlike static documentation generators (e.g., Docusaurus) or simple wiki tools, Moxie Docs is living and integrated. It doesn't just generate docs from code; it continuously re-indexes on every merge to stay current. It goes beyond passive documentation by actively enforcing standards on PRs and proactively serving context to AI agents, acting as an active layer of governance and intelligence rather than a passive repository.
2. Key Innovation: The core innovation is the bi-directional "single source of truth" architecture. The same living index powers both a human-readable workspace and machine-readable, scoped MCP context for agents. This ensures absolute consistency between what people read and what AI tools "know," with both paths being automatically updated via Git merges. The "Friday Cleanup" PR automation is also a unique approach to proactively combating documentation debt.

Frequently Asked Questions (FAQ)

1. How does Moxie Docs integrate with AI coding tools like Cursor or Claude? Moxie Docs provides repo context over the Model Context Protocol (MCP). Once your repository is indexed, tools like Cursor, Claude Code, and Copilot can pull conventions, known doc gaps, and search results directly into their context via MCP endpoints, ensuring they follow your project's standards without manual setup or codebase dumping.
2. Does Moxie Docs automatically modify or merge changes into my repository? No. Moxie Docs is designed for safety and review. It edits only pull request descriptions (not code) for alignment. All proposed documentation updates are delivered as separate, reviewable "Cleanup PRs" during the Friday Recap. Your team retains full control over merging via GitHub's native branch protection and code review processes.
3. What does the initial repository indexing process entail? The indexing process is secure and read-only. After you connect your GitHub repository, Moxie Docs scans your source code, test files, existing documentation, and commit history. It then builds a structured index of your architecture, conventions, and documentation patterns, which powers the searchable workspace and MCP context. The first index runs in minutes after setup.
4. Can I use Moxie Docs with public repositories or only private ones? Based on the provided pricing plans, Moxie Docs is currently tailored for private repositories, with plans (Solo and Team) specifying limits on active private repos. The tool is built to securely handle proprietary code and business logic for teams.
5. How does the "Friday Cleanup" feature work? Every Friday, Moxie Docs analyzes documentation gaps identified over the week from merged PRs and the living index. It then opens one or more docs-only pull requests that contain source-cited documentation updates. These PRs are clearly marked as automated and are placed into your review queue for approval, ensuring no unreviewed changes are automatically merged.