One signed binary. Every feature compiled in. Free to run. Install Crowkis →
← back to the Roost
guidesJune 2, 2026· 4 min read

Let Claude Code use the cache: Crowkis over MCP

One config block turns Crowkis into a tool an AI assistant can hold — check the cache, store the answer — over MCP, with the same trust pipeline as every other write.

The Model Context Protocol flips the integration: instead of your code calling the cache around the model, the assistant itself holds cache operations as tools. `crowkis mcp` speaks JSON-RPC over stdio and registers lookup, store, and stats surfaces any MCP-capable assistant can invoke mid-conversation.

register it with Claude Code
// .mcp.json  (or your client's MCP config)
{
  "mcpServers": {
    "crowkis": {
      "command": "crowkis",
      "args": ["mcp", "--host", "127.0.0.1", "--port", "6379"]
    }
  }
}

The mcp subcommand boots silently — the startup banner every other verb prints would corrupt a JSON-RPC stream, so it's the one quiet door. A hit returns the answer with its confidence; a miss returns a clean signal the assistant can act on rather than hallucinate around.

Crucially, MCP writes get no trust shortcut: an assistant's store request walks the same five-stage anti-poisoning pipeline as RESP and SDK writes, tracked in the ledger as its own source. An agent that stores garbage earns a higher bar automatically. The result is agents that remember as a behaviour — Claude Code checks before spending tokens, banks what it computes, and the whole team's assistants share the dividend.

One config block, and the cache your services already share becomes a tool your agents can hold too.