Skip to main content

MCPScanner

MCPScanner (formerly known as mcpry) is a command-line tool for analyzing MCP servers. It does the following:

  1. Discovers MCP Servers: It automatically searches for MCP server configuration files in well-known locations on the host system.
  2. Analyzes Tools and Resources: It connects to each discovered server to fetch the list of available tools and resources.
  3. Security Scanning with Pangea AI Guard: It uses the Pangea AI Guard service to scan the tools for malicious entities and prompts.
  4. Generates Reports: It creates a JSON report (default mcpscanner.json) containing the analysis results.
  5. Detects Changes: It can compare the current state of a server's tools with a previous report and display a diff if any changes are detected.
  6. Finds Similar Tools: It can identify tools with similar functionality.

Sample output

Installation

pip install -U mcpscanner

Configuration

Before using MCPScanner, you need to set the PANGEA_AI_GUARD_TOKEN environment variable to a Pangea API token that has access to the Pangea AI Guard service.

export PANGEA_AI_GUARD_TOKEN="pts_your_token_here"

Pangea domain

To use a Pangea domain other than the default aws.us.pangea.cloud, set the PANGEA_DOMAIN environment variable.

export PANGEA_DOMAIN="aws.us-west-2.pangea.cloud"

Usage

The primary command is scan, which runs the analysis.

mcpscanner scan

Options

Parameter Description Default
--input <PATH> The input file containing a previous report to compare against. mcpscanner.json
--output <PATH> The file where the new report will be saved. mcpscanner.json
--list-tools If set, the names of all tools for each MCP server will be listed in the output. False
--mcp-config-files <FILES> A list of files to discover MCP servers from. A list of well-known paths for different operating systems.
--similarity-threshold <FLOAT> The threshold (between 0.0 and 1.0) for two tools to be considered similar. 0.96
--syntax-theme <THEME> The syntax theme to use for displaying JSON diffs. github-dark
--poll-result-timeout Timeout (seconds) for polling AI Guard results. 30

Metadata

Release files for mcpscanner 0.5.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for mcpscanner 0.5.1
File Size Uploaded Provenance
mcpscanner-0.5.1.tar.gz 6.7 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for mcpscanner 0.5.1
File Interpreter ABI Platform
mcpscanner-0.5.1-py3-none-any.whl Python 3 none any Details

Total release size: 14.2 kB

Release files / mcpscanner-0.5.1.tar.gz

Download URL mcpscanner-0.5.1.tar.gz
Size 6.7 kB
Tags Source
SHA-256 checksum
How to use checksums
ae1098be0ba4f690be01d994bc1cccec649d8ebbc98e4ccf443c21201286af69
BLAKE2b-256 checksum
How to use checksums
a2222b9724f597445cc2b142b95ebcd6595823b9dc6212d0237dc864d1773ddf
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.1.0 CPython/3.13.7

Release files / mcpscanner-0.5.1-py3-none-any.whl

Download URL mcpscanner-0.5.1-py3-none-any.whl
Size 7.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
a57af1abe6bb5c16465898284497e7658021d5e8de2b8cafe7cadf5bdb9c6a39
BLAKE2b-256 checksum
How to use checksums
7ea3100276f6cabd059713410939bad8c21378d2abbbcac5e47f2cabc78db39c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.1.0 CPython/3.13.7

Security for mcpscanner 0.5.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page