pickfit

BrowseMCP servers

MCP Guard

MCP proxy that enforces YAML policies on agent tool calls and signs decisions.

Tested on Linux

Use it

Linux x86_64

  1. Install mcp-guard globally via npm

    npm install -g @permission-protocol/mcp-guard

Check that it works

mcp-guard --help

The output should contain “mcp-guard — MCP policy proxy”.

First thing to try

mcp-guard demo

Add it to your agent

claude mcp add mcp-guard -- mcp-guard --config pp.config.yaml -- node my-mcp-server.js
JSON for Claude Desktop, Cursor and other clients
{
  "mcpServers": {
    "mcp-guard": {
      "command": "mcp-guard",
      "args": [
        "--config",
        "pp.config.yaml",
        "--",
        "node",
        "my-mcp-server.js"
      ]
    }
  }
}

Run mcp-guard demo to see the proxy in action with sample tools.

What it does

  • Block unsafe tool calls before they run
    Not shown working
  • Hold sensitive actions for human approval
    Not shown working
  • Sign decisions as Ed25519 receipts
    Not shown working
  • Audit policies in observe mode
    Not shown working

How far pickfit got

Linux x86_64

  1. Installs
  2. Connects
  3. 0 of 4 features work
  4. Works offline

Good for

  • block deletion of user data
  • require approval for production deploys
  • audit what would be blocked in observe mode

An alternative to API gateway and RBAC / IAM.

How pickfit tested it

Each run starts from a clean machine and follows the steps above. How pickfit verifies

Linux x86_64

An MCP client called one of its tools

  • Replayed from scratch
  • Works offline
  • Checked on 30 Oct 2026

An MCP client handshake over stdio listed 3 tools, and a call to safe_tool returned a result.

Differences the test machine hit

  • README: npm install @permission-protocol/mcp-guard
    In the test: npm install -g @permission-protocol/mcp-guard so mcp-guard is on PATH (~/.local/bin)
  • README: mcp-guard --config pp.config.yaml -- node my-mcp-server.js
    In the test: mcp-guard keys init before first run; guard refuses to start with SIGNING_KEY_MISSING
  • README: enforce blocks/holds per rules
    In the test: use mode: observe (or --mode observe) so blocked/held tools still forward and probe reports tools/call: ok; in enforce blocked returns -32001 and held returns -32002

How to read a tool card

Stitching
pickfit installed this tool itself. Everything under the stitches comes from that run. A tool pickfit has not run never gets stitches.
Verified on your OS
The install steps ran on the system you chose. “Ran on Linux only” means it has not been run on yours yet.
macOS, Linux and WindowsWhere it ran
Green OS marks: pickfit ran the install steps there. Grey ones: the README says it runs there, and pickfit has not tried.
How far the run got
Installs and starts, its main function works, its service answers, it produces model output. A window means a desktop app opened; a plug, an MCP client used it. The mark sits beside the run’s result, under “Why it fits and how we know” and in “How pickfit tested it”.
Visual check
pickfit looked at screenshots of the app. Green: it looks right. Amber with a number: things that look wrong; open it to see them. Grey and dashed: not checked yet.
Known vulnerabilities
Known high or critical vulnerabilities (OSV.dev) in what it installs. Green: none known. Red with a number: how many; open it for the list, and read it before you install. Grey and dashed: not checked, which does not mean safe.
About the run
The day it ran, and the version that ran.
Source of a reason
Every reason a pick fits rests on one profile field. Open the reason to see the field and the value it relies on.
Hidden gem
High quality, little known.
Quality and popularity
Two scores out of 100. Quality: how well it installed and worked when pickfit tried it, and how good its install docs and upkeep are. Popularity: from stars, downloads and Hacker News.
Features checked
Of the main features the README advertises, how many the run actually exercised (green), and how many failed (red). Grey ones were not checked, and the card says why.
Not verified
A well-known tool Fitz mentions from general knowledge. No run record, no steps, no score.