requirements-review
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill analyzes untrusted content from the
docs/requirements/directory. It lacks explicit boundary markers to separate document content from instructions, creating a surface for indirect prompt injection where instructions embedded in requirements documents could attempt to bypass the audit criteria. Evidence Chain: 1. Ingestion points:docs/requirements/(SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Reading and writing files, and dispatching sub-agents. 4. Sanitization: Absent. - [NO_CODE]: The skill consists entirely of instructional text and logic definitions in Markdown and YAML, with no accompanying scripts or executable files, which reduces the risk of direct malicious code execution.
Audit Metadata