pm-clarify
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from local project files to extract and structure requirements, which represents a standard indirect prompt injection surface for this use case. \n
- Ingestion points: The skill reads from docs/01-需求调研/需求调研报告.md and docs/01-需求调研/创意方案库.md in Step 1. \n
- Boundary markers: Absent; the instructions do not specify delimiters to separate ingested content from instructions. \n
- Capability inventory: The skill utilizes Write (to generate documents), Bash (for file checks), and Agent (to delegate sub-tasks). \n
- Sanitization: No explicit validation or filtering of the file content is performed. \n- [COMMAND_EXECUTION]: The preamble utilizes the Bash tool to perform a local environment check. This check is limited to verifying the existence of a prerequisite file (using if [ ! -f ... ]) and providing status feedback to the user.
Audit Metadata