project-manager-senior
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions reference the execution of a local script
./qa-playwright-capture.shfor QA testing and automated screenshot capture. - [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The agent reads project specifications from
ai/memory-bank/site-setup.mdand extended methodology fromai/agents/pm.md. - Boundary markers: Absent; no delimiters are defined to isolate untrusted specification data from instructions.
- Capability inventory: The skill is designed to write task files to the local filesystem and execute shell scripts.
- Sanitization: Absent; the external specification content is not validated or escaped before being processed.
Audit Metadata