pm-docs
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local utility script
check-update.shfrom a relative path (../../check-update.sh) to perform version checks during the preamble stage. This is a vendor-related maintenance command. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes content from untrusted external documentation to generate summaries and plans.
- Ingestion points: Document files located in the
docs/01-需求调研/directory, including MVP plans and survey reports. - Boundary markers: None. The extracted information is directly interpolated into prompts for sub-agents without delimiters or instructions to ignore embedded instructions.
- Capability inventory: The skill uses
Writefor file creation,Agentfor sub-task delegation, andBashfor directory management and update checks. - Sanitization: The skill does not validate or sanitize the content of the documents it reads before passing it to the sub-agent prompts.
Audit Metadata