prd-to-steel-thread
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from external documents, which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill reads external PRD files and task lists from paths provided in the user arguments as described in the 'Read and assess the inputs' section of
SKILL.md. - Boundary markers: There are no instructions for the agent to use specific delimiters or to treat the ingested file content as non-executable data to prevent instruction overriding.
- Capability inventory: The agent has the authority to create and modify local files, specifically the
steel-thread.mdroadmap artifact. - Sanitization: The instructions explicitly require the agent to preserve PRD requirement and task descriptions 'verbatim' without providing a mechanism for sanitizing or validating the input data for potential injections.
Audit Metadata