Moxie Docs
Moxie Docs is an automated tool aimed at creating and maintaining documentation for GitHub repositories. It generates searchable documentation directly from the source code of a GitHub repository, making it easy to keep documentation up to date and consistent with the latest changes in the codebase.
Moxie Docs also integrates with AI development tools such as Claude, Cursor, Gemini, and Copilot. It identifies codebase conventions and spots existing documentation gaps when a repository is connected. The MCP server provides agents with tools for fetching codebase context, conventions, and docs - allowing you to save tokens by improving output and getting code right the first time.
The tool delivers source-cited documentation as a PR for review, ensuring that all PRs are kept in alignment automatically. Moxie Docs detects changes, or 'drifts', in the codebase when new merges are made and flags any corresponding documentation that needs to be updated.
Learn more
Unabyss
Unabyss is a universal context layer for AI tools that connects once and carries the same live, structured context into every AI a user works with. It pulls information from daily work sources such as Slack, Gmail, Notion, Google Calendar, GitHub, Linear, Google Drive, meeting tools, and more, then extracts, organizes, tags, and keeps that context current automatically. Instead of leaving knowledge trapped inside one chatbot or frozen in a manually maintained context file, Unabyss lets Claude, ChatGPT, Cursor, Codex, Gemini, Perplexity, OpenCode, VS Code, OpenClaw, and other MCP-compatible tools work from the same picture. Each AI can retrieve only the relevant slice of context, structured by topic, source, sensitivity, project, account, and other attributes, so users do not have to repeatedly explain their role, preferences, decisions, clients, repositories, or current priorities.
Learn more
Code Fundi
Code Fundi is a codebase intelligence platform that powers AI agents, engineering teams, and applications with a searchable map of software repositories. Instead of relying on grep, raw embeddings, or disconnected documentation, it ingests a repository from a URL or GitHub, maps files, functions, dependencies, and logic flows, removes boilerplate noise, and converts the result into a token-optimized representation ready for AI context windows. Its Blast-Radius Guard shows which files, services, and downstream behaviors may be affected before an agent changes code, helping teams prevent silent regressions and production failures. Universal Search runs sub-second semantic queries across one repository or hundreds of codebases, allowing users to find implementation patterns, compare architectures, inspect authentication or testing strategies, and trace how code works across projects.
Learn more
Asimov
Asimov is a code research agent that understands and works with complex enterprise codebases. Rather than focusing on code generation, it prioritizes codebase comprehension, a task that consumes up to 70% of developers’ time, by mapping relationships between code, architecture, and team decisions; maintaining institutional knowledge as engineers join and leave; and learning organically from team interactions and documentation. It indexes your entire development environment, including code repositories, architecture docs, GitHub threads, and Teams conversations, to build a persistent, cross‑cutting understanding of systems and to maintain context across architectural changes and team transitions. By using expanded context windows instead of traditional retrieval methods, Asimov can dynamically reference any part of a codebase during reasoning, enabling more accurate synthesis across disparate components.
Learn more