codehealth-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to configure and run the @codescene/codehealth-mcp package using npx. This package is associated with a well-known service (CodeScene).
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes external code files and git state which are untrusted ingestion points. Evidence: 1. Ingestion points: The code_health_review, code_health_score, and analyze_change_set tools read local repository files and git context. 2. Boundary markers: Absent; there are no instructions for the agent to ignore or delimit instructions found within the analyzed source code. 3. Capability inventory: The skill reads file contents and git history; the resulting health scores and reviews are used by the agent to make decisions about refactoring and committing code. 4. Sanitization: Absent; the skill does not specify filtering or escaping of the code content before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 03:39 PM
Security Audit — agent-trust-hub — codehealth-mcp