product-doc-audit

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill reads and processes untrusted documentation files (PRDs, specs, design docs) provided by the user. While it is limited to creating a report, instructions embedded within those documents could theoretically influence the agent's output logic.
  • Ingestion points: Target document directory or file index (SKILL.md).
  • Boundary markers: The skill does not employ specific delimiters to isolate the text being audited from the agent's instruction context.
  • Capability inventory: Writing and overwriting the PRODUCT-DOC-AUDIT.md report file (SKILL.md).
  • Sanitization: No explicit sanitization or filtering of the ingested document content is described.
  • [SAFE]: The skill includes hard constraints to prevent the modification of source documents and explicitly restricts its output to a single file within the audited directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:55 PM
Security Audit — agent-trust-hub — product-doc-audit