MCP Registry

github-mcp

by seeygo
3adae292-17ea-5511-999e-2f023c93e1f9 | Store v0.0.7
0/ 100
MINIMAL risk
AI review verdict
No high-risk signal observed

Based on the RiskyPlugins AI security review of the observed evidence.

No individual score drivers were recorded for this analysis.

Analysis record

Analysed
7 months ago 18 April 2026
Version
v0.0.7
Detail
Full list

No findings

The analyzer reported nothing for this version.

No findings

The analyzer reported nothing for this version.

AI Security Report

AI Security Review

Evidence context: threat category pup; evidence quality weak.

The github-mcp package presents a risk score of 87.14 primarily due to indicators of compromise (IoCs) and malware-signature matches. However, a detailed analysis reveals these findings are characteristic of a standard Node.js distribution bundle rather than malicious activity.

Tool Poisoning Assessment: No tool-poisoning findings were detected. This is a strong positive signal, indicating the tool definitions and descriptions do not contain hidden directives aimed at manipulating the AI agent.

Network & Credential Scope: The package includes a node_modules directory (48 IoCs, 32 malware signatures). IoC counts are notoriously inflated in bundled dependencies due to minified code fragments and internal infrastructure calls (e.g., npmjs.org, cloudflare.com) being misidentified as threats. Without evidence of targeted credential harvesting (e.g., process.env exfiltration to unknown domains) or shadow tool behavior, the network findings are likely standard library operations for GitHub OAuth or API calls.

Conclusion: While the bulk of findings results from bundled dependencies, the absence of MCP-specific attack vectors (tool poisoning, exfiltration) suggests this is a legitimate tool triggering generalized scanner noise. The primary risk is inherent in the package's broad permissions, not malice.

Key Reasons

  • Zero tool-poisoning findings detected despite high risk score
  • IoC and signature findings are consistent with bundled node_modules artifacts
  • No evidence of targeted credential harvesting or exfiltration architecture

False Positive Considerations

  • Bundled dependencies in node_modules causing IoC inflation
  • Malware-signature rules matching minified vendor code

Reviewed 2026-04-13; recommended action: suppress false positive; model confidence 72%.

Source Code Viewer

Source Code Not Available

Source code is not available for this version of the extension.

About This Extension

GitHub MCP Server with OAuth support for Issue management

Frequently Asked Questions