pm-discovery
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines an automated pipeline that ingests user-provided input to generate project documentation, which creates a potential surface for indirect prompt injection.
- Ingestion points: The
{feature}argument is passed through thetask-templateand shared across multiple agents (pm-lead,pm-discovery,pm-strategy, etc.) in the workflow. - Boundary markers: The skill definition does not specify explicit boundary markers or instructions to ignore potential instructions embedded within the
{feature}text string. - Capability inventory: The workflow utilizes powerful tools including
Bash,Write,Edit, andGlobto analyze project structure and write documentation to the file system. - Sanitization: No explicit sanitization or validation logic is defined for the user-supplied input before it is interpolated into agent tasks or file paths.
Audit Metadata