mcp-server-id-mapping

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill references spawning a Node.js process (node legiox-mcp-server.js --test) as a health check within a packager CI environment. This is presented as a development pattern for verification.
  • [INDIRECT_PROMPT_INJECTION]: The instructions involve reading external tool descriptor JSON files from paths like mcps/*/tools/*.json to ensure schema compatibility before tool invocation. This identifies a surface where untrusted data could be ingested into the agent context.
  • Ingestion points: External tool schemas and metadata files located in mcps/*/tools/*.json and ~/.cursor/projects/*/mcps/.
  • Boundary markers: Not specified in the instructions.
  • Capability inventory: Mentions the use of CallMcpTool and node subprocess execution for testing.
  • Sanitization: The skill focuses on ID mapping and schema validation rather than input sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 01:40 PM
Security Audit — agent-trust-hub — mcp-server-id-mapping