visual-review

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local project scripts (scripts/skills-browser.sh) and standard development tools (Playwright, bun run dev) to perform visual reviews. These are intended functionalities for a developer-oriented tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests external data from git diff and potentially interactive browser content during UI reviews.
  • Ingestion points: Reads git diff --name-only HEAD, route hints, and content from the browser via Playwright in SKILL.md and REFERENCE.md.
  • Boundary markers: No specific delimiters or warnings against embedded instructions are specified in the provided files.
  • Capability inventory: The skill is capable of executing shell commands via Playwright and local scripts as described in SKILL.md.
  • Sanitization: No explicit sanitization or filtering of external content is mentioned.
  • [DATA_EXPOSURE]: The skill captures an environment fingerprint including User Agent, platform, and viewport settings to facilitate UI bug reproduction. This data is intended for inclusion in pull request reports as context for reviewers and does not involve sensitive credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 05:07 PM
Security Audit — agent-trust-hub — visual-review