slice-the-spec
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data (Product Requirements Documents and codebase files) which may contain malicious instructions designed to influence the agent's output or behavior.
- Ingestion points: Step 1 ('Locate and understand the PRD') and Step 2 ('Explore the codebase') in
SKILL.mdinvolve reading potentially attacker-controlled content from the repository or external documents. - Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions to isolate the content of the PRD from the agent's logic.
- Capability inventory: The skill performs file reads and exploration, and generates GitHub-ready issue text.
- Sanitization: There are no explicit instructions for the agent to sanitize or escape the content of the PRD before including it in the generated issues.
Audit Metadata