Skip to content
MCP server

LynxPrompt

Search, read and manage AI agent instruction blueprints such as AGENTS.md on LynxPrompt

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.3.2
Listing updated
2 days ago

About LynxPrompt

LynxPrompt is an open-source MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.

LynxPrompt stores AI configuration files (blueprints) such as AGENTS.md and CLAUDE.md, with a public marketplace and groupings called hierarchies for monorepos. This server lets an assistant browse the marketplace, list your own blueprints and hierarchies, read any blueprint you can see, and create, update or delete them. It points at the hosted instance by default or at your own self-hosted LynxPrompt.

Updates are guarded: update_blueprint takes the content_checksum returned by get_blueprint as expected_checksum, so it cannot silently overwrite a change made elsewhere. Setting LYNXPROMPT_READ_ONLY registers only the read tools, and read-only resources expose blueprints, hierarchies and the current user.

Searching the marketplace works without a token, while everything else needs an API token created in your LynxPrompt settings. It runs over stdio from npm, Docker or Go binaries, or over stateless HTTP with origin checks and bearer auth, and is licensed GPL-3.0-or-later.

Install

Run in a terminal

claude mcp add lynxprompt-mcp --env LYNXPROMPT_TOKEN='<LYNXPROMPT_TOKEN>' -- npx -y lynxprompt-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "lynxprompt-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "lynxprompt-mcp"
      ],
      "env": {
        "LYNXPROMPT_TOKEN": "<LYNXPROMPT_TOKEN>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "lynxprompt-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "lynxprompt-mcp"
      ],
      "env": {
        "LYNXPROMPT_TOKEN": "<LYNXPROMPT_TOKEN>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "lynxprompt-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "lynxprompt-mcp"
      ],
      "env": {
        "LYNXPROMPT_TOKEN": "<LYNXPROMPT_TOKEN>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "lynxprompt-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "lynxprompt-mcp"
      ],
      "env": {
        "LYNXPROMPT_TOKEN": "<LYNXPROMPT_TOKEN>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "lynxprompt-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "lynxprompt-mcp"
      ],
      "env": {
        "LYNXPROMPT_TOKEN": "<LYNXPROMPT_TOKEN>"
      }
    }
  }
}

Configuration

  • LYNXPROMPT_TOKENLynxPrompt API token (lp_...), created under Settings > API tokens. Needed for everything except search_blueprints.Required
  • LYNXPROMPT_URLLynxPrompt instance URL; https:// unless the host is loopback.
  • LYNXPROMPT_READ_ONLYSet to true to register only the read tools (no create, update or delete).
  • LYNXPROMPT_TOKENLynxPrompt API token (lp_...), created under Settings > API tokens. Needed for everything except search_blueprints.Required
  • LYNXPROMPT_URLLynxPrompt instance URL; https:// unless the host is loopback.
  • LYNXPROMPT_READ_ONLYSet to true to register only the read tools (no create, update or delete).

Tools this server provides (10)

Taken from the project documentation; the running server may differ.

search_blueprints
Searches the public marketplace.
list_blueprints
Lists your blueprints.
get_blueprint
Reads a blueprint with its checksum.
list_hierarchies
Lists your hierarchies.
get_hierarchy
Reads one hierarchy with its blueprints.
create_blueprint
Creates a blueprint.
update_blueprint
Updates a blueprint using an expected checksum.
delete_blueprint
Permanently deletes a blueprint.
create_hierarchy
Creates a hierarchy.
delete_hierarchy
Deletes a hierarchy.

What it does

Marketplace search
Search public blueprints by query, category and tags.
Own blueprints
List, read, create, update and delete your own.
Hierarchies
Group blueprints for monorepos.
Checksum-guarded updates
Prevents overwriting edits made elsewhere.
Read-only mode
Hides the write tools with one setting.
Self-hosting support
Point it at your own LynxPrompt URL.

Good for

  • Find a ready-made CLAUDE.md for a framework
  • Publish a blueprint from an assistant
  • Update a team's shared agent instructions safely
  • Organise blueprints for a monorepo into a hierarchy

Pricing

Open source

GPL-3.0-or-later

LynxPrompt FAQ

Do I need an account?

Marketplace search works without a token; all other tools need an API token from Settings, API tokens.

Can it overwrite my work by mistake?

update_blueprint requires the checksum from get_blueprint so it does not silently overwrite edits.

Can I use my own instance?

Yes, set LYNXPROMPT_URL to a self-hosted LynxPrompt.

Can I make it read-only?

Yes, set LYNXPROMPT_READ_ONLY to true.

Is LynxPrompt free?

Yes. LynxPrompt is open source, so you can use and self-host it for free.

How do I add LynxPrompt to Claude Code?

Run this command in your terminal: claude mcp add lynxprompt-mcp --env LYNXPROMPT_TOKEN='<LYNXPROMPT_TOKEN>' -- npx -y lynxprompt-mcp

What are the best alternatives to LynxPrompt?

The closest alternatives to LynxPrompt are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to LynxPrompt

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools