Persistent knowledge graph and MCP server for AI workflows with git-tracked project context.
Persistent knowledge graph and MCP server for AI workflows with git-tracked project context.
It's a persistent knowledge graph for AI-assisted workflows. It indexes your conversations, projects, and intellectual work so that Claude Code, Codex, Gemini CLI, Google Antigravity, OpenCode, Cursor, and other MCP-capable agents never start a session blind. Available as a free MCP plugin or standalone CLI.
Compare MCP servers by category, or browse hosted MCP servers if you prefer an endpoint over a local process.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| payments-refactor | 2 unread (focus: Extract billing service) |
| mcp | kin-mcp MCP server (for Claude Code, Codex, Gemini, Antigravity, OpenCode, Cursor, etc.) |
| llm | Anthropic-powered extraction (kin learn, kin ask) |
| vectors | sqlite-vec for semantic similarity search |
| reminders | Natural-language time parsing for kin remind |
| all | Everything above |
| full | ~4000 tokens |
| abridged | ~1500 tokens |
| summarized | ~750 tokens |
| executive | ~200 tokens |
| index | ~100 tokens |
{ "mcpServers": { "kindex": { "command": "kin-mcp", "args": [] } } }Configuration as documented by the project. Restart the client after saving.
| Variable | Description | Required |
|---|---|---|
| CLAUDE_SESSION_ID | Configuration value read at startup. | Optional |
| CODEX_SESSION_ID | Configuration value read at startup. | Optional |
| OPENCODE_SESSION_ID | Configuration value read at startup. | Optional |
| CURSOR_SESSION_ID | Configuration value read at startup. | Optional |
| KIN_AGENT_ID | Configuration value read at startup. | Optional |
Store entities, relations and observations in a local knowledge graph with the reference Memory MCP server.
A structured scratchpad for hard problems — stepwise reasoning with revisions, branches and visible logic.
The recommended way to use this MCP server is through FastMCP, which provides better performance and easier configuration.
Persistent memory for AI agents — history, context, semantic search, queues. 17 tools, stdio/SSE.
Persistent memory for AI assistants: store, search, and connect knowledge across conversations.
Docsie MCP lets you create, manage, and publish documentation, convert videos to docs, run compliance analysis, and search your knowledge base - all