# toolprint by JEStats

> toolprint by JEStats is an open-source MCP server for security. Pins MCP tool definitions in a lockfile and flags rug-pulls, poisoning and leaked secrets. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, API, macOS, Windows, Linux
- **Website:** https://jestatsio.github.io/toolprint/
- **Repository:** https://github.com/jestatsio/toolprint
- **MCP Registry name:** io.github.jestatsio/toolprint
- **Categories:** security, developer tools
- **Listing updated:** 2026-11-15

## About

toolprint treats MCP trust like a dependency lockfile. It fingerprints every tool, prompt, resource and skill definition into a committed toolprint.lock, so a rewritten description appears as a diff in a pull request instead of silently changing what an agent reads.

A scan runs four checks. A pinned definition that changes fails with a before and after diff. Eleven injection signals look for tool poisoning, and two independent high-severity signals on one capability escalate it to critical. Eighteen credential formats are detected and always redacted in output. Targets can be a config file, a URL, an npx spec or a raw command, and --all-clients covers every agent client on the machine. SKILL.md bundles are scanned too.

The engine ships as a CLI, a GitHub Action with SARIF output and sticky PR comments, and an MCP server started with toolprint serve. That server exposes scan (read-only) and pin, available as plugins for Claude Desktop, Claude Code and Codex. It runs locally with no account, and the maintainers note that their packaging is community-distributed rather than officially endorsed.

## Features

- **Lockfile pinning:** Commit toolprint.lock so changes appear as reviewable diffs.
- **Rug-pull detection:** A changed definition fails the scan with a before and after view.
- **Poisoning signals:** Eleven injection checks across capability text and skill bodies.
- **Secret redaction:** Eighteen credential formats are detected and masked.
- **CI integration:** GitHub Action with SARIF, PR comments and a baseline mode.

## Good for

- Review MCP server changes in a pull request
- Audit every agent client config on a developer machine
- Block a build when hidden prompt injection is found
- Let an agent verify tools before it trusts them

## Install

### Claude Code (Terminal)

```sh
claude mcp add toolprint --env TOOLPRINT_BEARER='<TOOLPRINT_BEARER>' -- npx -y toolprint
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "toolprint": {
      "command": "npx",
      "args": [
        "-y",
        "toolprint"
      ],
      "env": {
        "TOOLPRINT_BEARER": "<TOOLPRINT_BEARER>"
      }
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "toolprint": {
      "command": "npx",
      "args": [
        "-y",
        "toolprint"
      ],
      "env": {
        "TOOLPRINT_BEARER": "<TOOLPRINT_BEARER>"
      }
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "toolprint": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "toolprint"
      ],
      "env": {
        "TOOLPRINT_BEARER": "<TOOLPRINT_BEARER>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "toolprint": {
      "command": "npx",
      "args": [
        "-y",
        "toolprint"
      ],
      "env": {
        "TOOLPRINT_BEARER": "<TOOLPRINT_BEARER>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "toolprint": {
      "command": "npx",
      "args": [
        "-y",
        "toolprint"
      ],
      "env": {
        "TOOLPRINT_BEARER": "<TOOLPRINT_BEARER>"
      }
    }
  }
}
```

## Tools

- `scan`: Read-only scan of MCP servers or skills for drift, poisoning and secrets.
- `pin`: Updates toolprint.lock after review.

## FAQ

### Does it need an account?

No. The README says it runs locally with no account.

### What does the MCP server expose?

Two tools: scan, which is read-only, and pin, which updates the lock after review.

### How do I try it?

Run npx toolprint pin on your MCP config, then npx toolprint scan.

### Does it scan skills?

Yes. npx toolprint scan --skills covers SKILL.md bundles.

### Is toolprint by JEStats free?

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

### How do I add toolprint by JEStats to Claude Code?

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

### What are the best alternatives to toolprint by JEStats?

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

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Concord MCP](https://ai-powered.com/mcp/concord-mcp): Concord MCP is a freemium MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/toolprint-by-jestats
