eval-sys/mcpmark

MCPMark is a comprehensive, stress-testing MCP benchmark designed to evaluate model and agent capabilities in real-world MCP use.

441 starsPythonApache-2.0benchmarkmcp-serversmcpmarkeval-sys

SCANNED65/100

Scanned 2026-07-10 · deepseek v1

Dimension breakdown

Security

65/100

9 findings

Quality

78/100

No findings

License

95/100

No findings

Completeness

15/100

No findings

Findings (9)

HIGH: 3MEDIUM: 3LOW: 2INFO: 1

HIGHsecurityHardcoded API key in README example
HIGHsecurityPotential SSRF via user-controlled URL in NotionStateManager
HIGHsecurityUnsafe file operations with path traversal risk
MEDIUMsecurityInsecure storage of authentication state
MEDIUMsecurityPotential command injection via subprocess calls

Detailed file:line citations + recommended fixes are visible to the maintainer of this repo after claiming the listing.

Maintainer of this repo?

Claim this listing in 30 seconds with GitHub OAuth. You'll see detailed findings, get notified about matched bounties, and can trigger re-scans on demand.

Claim listing

Embed the badge.

Drop it into your README. Every view backlinks to this verification page.

Markdown

[![Archimedes Trust](https://archimedes.market/api/badge/eval-sys/mcpmark.svg)](https://archimedes.market/r/eval-sys/mcpmark)

HTML

<a href="https://archimedes.market/r/eval-sys/mcpmark"><img src="https://archimedes.market/api/badge/eval-sys/mcpmark.svg" alt="Archimedes Trust Verified"/></a>