browser-verification
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted content from browser DOM snapshots and console messages to verify UI implementations.
- Ingestion points: Untrusted data enters the agent context through
mcp__plugin_playwright_playwright__browser_snapshotand console message tools as defined inreferences/verification-flow.md. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between the page content and the verification instructions.
- Capability inventory: The skill uses browser automation tools (navigate, click) and has the ability to write screenshot files to the
.ohno/screenshots/directory. - Sanitization: The skill lacks defined sanitization or validation logic for the content retrieved from the browser environment.
Audit Metadata