ecc-claude-ra-qm-team-fda-consultant-specialist

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its dependency on external reference snapshots for tool execution.
  • Ingestion points: The skill instructions specify reading a reference snapshot from the path provided in references/upstream-path.txt (upstream-import/claude-skills/references/ra-qm-team/fda-consultant-specialist).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the imported reference content were identified.
  • Capability inventory: The skill is designed to translate external guidance into 'tool-backed steps' and execute them, providing a mechanism for untrusted content to influence agent actions.
  • Sanitization: No sanitization or validation routines for the external content were described in the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 04:46 PM
Security Audit — agent-trust-hub — ecc-claude-ra-qm-team-fda-consultant-specialist