fittings-ctgov-eligibility-screen

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [UNVERIFIABLE_DEPENDENCIES]: The skill instructions direct the user to download a JavaScript file (call.mjs) from the vendor's domain (fittings.sh) using curl and execute it with node. While the source is the skill's own vendor, this pattern introduces a dependency that is fetched at runtime rather than being bundled with the skill.
  • [DYNAMIC_EXECUTION]: The skill relies on the execution of the downloaded call.mjs script to interact with the API and settle payments, representing a runtime execution of external code.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes clinical trial data, including inclusion and exclusion criteria, from the ClinicalTrials.gov API.
  • Ingestion points: Trial criteria text retrieved from the ctgov.eligibility-screen endpoint.
  • Boundary markers: No specific delimiters are mentioned in the instructions to isolate trial data from the agent's control flow.
  • Capability inventory: The agent performs analysis and decision-making (screening) based on the ingested trial criteria.
  • Sanitization: The documentation does not describe explicit sanitization or filtering of the content retrieved from the external registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 10:31 PM