hl-spec
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from feature descriptions, interview summaries, and external brief files which could contain malicious instructions designed to influence the behavior of the agent or downstream agents reading the output.
- Ingestion points: SKILL.md reads user-described features and interview summaries; references/prd.md accepts idea text and brief path.
- Boundary markers: The instructions do not specify the use of delimiters to isolate untrusted input.
- Capability inventory: The skill is authorized to read existing file structures and write multiple files to the local filesystem.
- Sanitization: There is no evidence of input validation or escaping for the ingested content.
Audit Metadata