Semantic code retrieval over a local graph of entities, relationships, history, and provenance.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag — we're steadily working through the catalog.
💡 Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Kin.
The Kin MCP server turns a code repository into a graph-native project model for people and AI agents. The model includes functions, types, other parsed entities, relationships between them, source content, and versioned history. These records are stored as repository state rather than as a separate search index layered on top of another repository.
Kin can import an existing Git repository without modifying tracked files. Initialization adds a .kin directory and a line to .git/info/exclude. Git history reachable from branches and tags is imported, while uncommitted and untracked changes remain outside the imported history. Kin also has its own commits, branches, merges, diffs, and history, and can work without Git in supported workflows.
Use the Kin MCP server when an agent needs context around a symbol or change: recorded callers, nearby code, related entities, or possible downstream effects. Results should be checked against the source, compiler, tests, and review because graph coverage is not complete.
The CLI and MCP server query the same local graph. Typical command-line exploration begins with kin locate, followed by kin refs, kin trace, or kin impact for a known entity. refs returns recorded references, trace adds nearby context, and impact follows possible effects through the graph.
Semantic retrieval uses a local model of approximately 523 MB. Before embeddings are available, kin locate falls back to lexical and graph signals and reports that vector coverage is missing. Running kin embed downloads the model when needed and builds the index. Local imports, storage, and queries run on the user’s machine; installation and the initial model download require network access.
Kin can also produce an advisory review report for an explicit commit range. The report uses graph-derived impact and supporting evidence and returns PASS, NEEDS ATTENTION, or WOULD BLOCK. It does not block merges or modify graph state.
Install on macOS, Linux, or WSL2 with Node.js 20 or newer:
The shell installer is available through get.kinlab.ai or get.kinlab.dev. Native Windows x64 has a PowerShell installer, but it does not connect AI coding tools; WSL2 is the recommended Windows path. No native Windows ARM64 build is published.
After installation, run kin init . in a full clone, or use kin clone with a repository URL. kin setup connects supported AI tools it finds after confirmation, and kin setup status shows current connections. Use kin setup --intent editor for VS Code. The setup process can also configure Codex CLI and Grok CLI when permitted.
The Kin MCP server is available to Claude Code, Codex, Cursor, Gemini, and other MCP clients. Client-specific setup instructions are required for some integrations.
Kin is in public beta. Shallow clones, submodules, Git LFS, sparse checkouts, active hooks, and unfinished merges or rebases are not supported during import. Supported languages receive parsed entities and relationships, but unsupported files remain available as content and history. An empty result does not establish that no caller or dependency exists.
Compatibility differs by platform, and the deprecated kin-vfs filesystem projection is not included in release archives. Review results are advisory, and authorship is declared rather than independently verified. Back up Kin-only state such as commits, branches, reviews, and specifications.
Factual signals from GitHub, npm, and our automated checks — not a rating.
No reviews yet — be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/kin)<a href="https://allmcps.com/mcp/kin"><img src="https://allmcps.com/api/badge/kin?style=directory" alt="Kin on AllMCPs" /></a>