agency-senior-project-manager
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill is instructed to execute a local shell script,
./qa-playwright-capture.sh, for automated quality assurance and screenshot capturing. - Evidence: Found in
SKILL.md:Include Playwright screenshot testing: ./qa-playwright-capture.sh http://localhost:8000 public/qa-screenshots. - [EXTERNAL_DOWNLOADS]: The skill suggests fetching placeholder images from established external providers.
- Evidence: Found in
SKILL.md:Images from approved sources (Unsplash, https://picsum.photos/). - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it is designed to read and process project specifications from a local file which could potentially contain adversarial instructions.
- Ingestion points:
ai/memory-bank/site-setup.md(viaSKILL.md). - Boundary markers: None identified in the instructions for separating untrusted specification content from agent instructions.
- Capability inventory: The agent has the capability to write to the local file system (
ai/memory-bank/tasks/) and execute local scripts (./qa-playwright-capture.sh). - Sanitization: No explicit sanitization or validation of the specification file content is mentioned.
Audit Metadata