HotKV MCP Server

A Model Context Protocol server for HotKV, a RESP2/RESP3-compatible in-memory data store with built-in AI and LLM-serving features (semantic and prompt caching, agent memory, RAG, feature store, rate limiting, time series and more). It lets an AI assistant or agent (Claude, Cursor, VS Code, your own) read and write keys, keep conversation memory and notes, log events, store prompt templates and search documents in a HotKV server.

• A fixed set of 28 tools. There is no tool that runs an arbitrary command, so an agent cannot flush a database, change the server's configuration or read the keys of other tenants.
• --read-only removes every write tool, so the agent cannot even see them.
• --namespace confines the agent to one prefix: it can only use and list names under it.
• Results are capped (--max-value-bytes, --max-items), and a cut value is flagged truncated, so a huge value cannot flood the model's context.
• stdio for local clients, streamable HTTP for shared use.

Written against HotKV server 0.3.0 and the official hotkv-client Python package.

Install

pip install hotkv-mcp

or run it without installing, with uv:

uvx hotkv-mcp --url hotkv://localhost:6379

Python 3.10 or newer.

Connect a client

The server speaks stdio by default, so a client launches it as a subprocess.

Claude Code

claude mcp add hotkv -- uvx hotkv-mcp --url hotkv://localhost:6379

Claude Desktop, Cursor and other clients that read an mcpServers file

{
  "mcpServers": {
    "hotkv": {
      "command": "uvx",
      "args": ["hotkv-mcp", "--read-only"],
      "env": { "HOTKV_URL": "hotkv://:your-password@localhost:6379" }
    }
  }
}

Keep the password in HOTKV_URL (or the client's secret store) and not on the command line, where other users of the machine can see it. Use hotkvs:// for a TLS connection.

Options

Command line options win over the environment.

Namespaces are plain prefixes: give each agent a namespace that is not itself a prefix of another (agent-a and agent-b, not agent and agent:b), or the shorter one can reach the longer one's names.

Tools

Each tool says in its MCP annotations whether it only reads, writes, or deletes, so clients can ask for approval before the destructive ones. With --read-only the tools that write (set, delete, hash_set, list_push, the memory_ tools that add or clear, prompt_save, rag_create, rag_ingest) are not registered at all.

The memory, prompt and document search tools use HotKV's AI engines, which need a HotKV license that includes them. Without one the tool returns "This HotKV server is not licensed for that feature" and the key tools keep working. rag_search ranks by keyword relevance, so it needs no embedding model.

Running over HTTP

hotkv-mcp --transport streamable-http --host 127.0.0.1 --port 8000

The server listens on /mcp. It does not authenticate callers. Keep it on localhost, or put it behind a reverse proxy that does, and give it its own --namespace and --read-only where you can. Starting it on another address prints a warning.

Development

pip install -e ".[dev]"
python -m pytest tests            # unit tests; no server needed
HOTKV_TEST_URL=hotkv://127.0.0.1:6379 python -m pytest tests   # also the integration tests

The integration tests need a HotKV server licensed for the AI engines and create and delete keys with a random prefix. Run them against a throwaway server.

Contributing

Bug reports and pull requests are welcome. See CONTRIBUTING.md.

Support

• Bugs and feature requests: GitHub issues
• Questions and commercial support: support@hotkv.com
• Security vulnerabilities: see SECURITY.md

License

Licensed under the Apache License, Version 2.0: see LICENSE and NOTICE.

Trademarks and affiliation

"HotKV" is a trademark of HotKV Ltd; the license does not grant rights to use it. The HotKV server is a separate commercial product, and its enterprise commands need a HotKV license.

HotKV is an independent product of HotKV Ltd. It speaks the RESP protocol and implements many Redis commands so that existing tools and client habits carry over, but it is not Redis, Valkey, Dragonfly or KeyDB, and this package is built for and tested against HotKV. HotKV Ltd is not affiliated with, endorsed by or sponsored by Redis Ltd., the Valkey project, DragonflyDB or KeyDB. Redis is a registered trademark of Redis Ltd. Valkey, Dragonfly, KeyDB and all other product and company names are trademarks of their respective owners; they are used here only to describe protocol and command compatibility. "Model Context Protocol" and "MCP" refer to the open protocol published at modelcontextprotocol.io.