mfds-food-safety

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions direct the agent to download the @nomadamas/k-skill package from the npm registry using npx. This package is a vendor-owned resource.
  • [REMOTE_CODE_EXECUTION]: The use of npx to execute commands from a remote package constitutes a remote code execution pattern. This is the intended mechanism for the vendor's ecosystem to provide up-to-date instructions and helper scripts.
  • [COMMAND_EXECUTION]: The skill provides specific shell commands using npx for the agent to execute in order to fetch instructions and run data lookup scripts.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: scripts/mfds_food_safety.py retrieves records from external food safety APIs via k-skill-proxy.nomadamas.org.
  • Boundary markers: No explicit delimiters are used to separate external data from the agent's internal prompt context.
  • Capability inventory: The skill is capable of shell command execution (npx) and network requests (urllib.request).
  • Sanitization: The helper script includes a summarize_text function that removes HTML tags and unescapes content, providing basic sanitization of the retrieved data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 07:51 AM