ns-proto-creator
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by processing untrusted content from external websites during the capture phase.
- Ingestion points: In
references/capture-playwright.md, the agent is instructed to navigate to external URLs and capture screenshots, accessibility snapshots, and structural notes. - Capability inventory: The skill has the capability to write files to the workspace (under
prototype/anddocs/) and interact with browser sessions via Playwright tools (click,fill,navigate). - Boundary markers: The instructions do not specify any delimiters or safety markers to isolate captured website content from the agent's internal instructions.
- Sanitization: There is no mention of sanitizing or validating text extracted from live websites before using it to generate prototype code or AS-IS documentation.
Audit Metadata