Skip to content
MCP server

Krewbill

Approved shared notes per client that Claude, ChatGPT and Cursor read before answering

Install
Pricing
Free
Platforms
Web, API
Runs
Remotely
Version
0.1.0
Listing updated
3 days ago

About Krewbill

Krewbill is a free MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Krewbill addresses the habit of re-pasting the same tone rules and background into every new chat. Each client, project or course gets a space of approved notes, and any assistant that supports remote MCP connectors reads those notes before it answers.

Learning goes through a person. During a conversation the assistant can suggest a note, which waits in an approval queue next to any similar existing note; nothing becomes shared memory until someone approves it. Access uses one key per tool, limited to the spaces you choose, shown once and revocable at any time. The notes are meant as reference, and the suggested instruction tells the assistant to treat them as background, not commands.

Setup is three steps: make a key under Connections, add the address as a custom connector (Claude, ChatGPT) or MCP server (Claude Code, Cursor), then ask what it knows about a space. It needs an AI plan that allows custom connectors, and the team has tested only a few tools.

Install

Run in a terminal

claude mcp add --transport http krewbill-mcp https://www.krewbill.com/api/mcp

Add to Settings → Connectors → Add custom connector

https://www.krewbill.com/api/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "krewbill-mcp": {
      "url": "https://www.krewbill.com/api/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "krewbill-mcp": {
      "type": "http",
      "url": "https://www.krewbill.com/api/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "krewbill-mcp": {
      "serverUrl": "https://www.krewbill.com/api/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "krewbill-mcp": {
      "type": "http",
      "url": "https://www.krewbill.com/api/mcp"
    }
  }
}

Tools this server provides (2)

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

get_space_context
Reads the approved notes for a space as background.
suggest_memory
Proposes a new note that waits for human approval.

What it does

Space per client
Keeps notes separate for each client, project or course.
Approval queue
Assistant suggestions wait for a person before they become shared memory.
Scoped keys
One key per tool limited to chosen spaces, revocable at any time.
Works across assistants
Reads from Claude, ChatGPT, Cursor and other remote-MCP tools.
Duplicate-aware review
Suggestions appear beside similar notes.

Good for

  • Keep a client's tone and rules consistent across teammates
  • Stop re-explaining project background in new chats
  • Capture corrections from a conversation as reviewed notes
  • Give each agency client its own isolated AI memory

Pricing

Free

Free early access until Dec 30

Prices change. Check krewbill.com before you buy.

Krewbill FAQ

Who approves new notes?

A person. Suggestions wait in an approval queue and nothing goes live without approval.

Which tools work with it?

Any AI tool with remote MCP connectors; Claude, ChatGPT and Cursor are named, and only a few have been tested.

What does it cost?

Early access is free until Dec 30; later pricing is not given.

What if a key leaks?

Revoke it in Connections; each tool has its own key.

Is Krewbill free?

Yes. Krewbill is free to use.

How do I add Krewbill to Claude Code?

Run this command in your terminal: claude mcp add --transport http krewbill-mcp https://www.krewbill.com/api/mcp

What are the best alternatives to Krewbill?

The closest alternatives to Krewbill are Codebase Memory, XERJ, Ha MCP, Claude Code Ultimate Guide and Nautilus Compass Memory.

Alternatives to Krewbill

  1. 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
  2. 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
  3. Ha MCP

    MCP server

    MCP server for managing and automating Home Assistant from AI assistants

    Runs locally
    • 8 tools
    • Open source
    • 7.9K GitHub stars
    • Productivity
  4. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  5. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  6. Concord MCP

    MCP server

    Local-first MCP server for messaging and shared work-state between coding agents

    Runs locally
    • Freemium
    • 1.3K GitHub stars
    • Developer tools