pr-visual-explainer

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted content from pull requests (diffs, tests, and PR prose) to generate a visual walkthrough. This presents a surface for indirect prompt injection where malicious instructions could be embedded in the PR data to influence the agent's explanation. However, the skill includes explicit instructions to treat all PR-derived content as untrusted and mandates HTML entity escaping for text and attributes.
  • [DATA_EXPOSURE]: The instructions explicitly forbid the inclusion of secrets, credentials, tokens, private URLs, or personal data in the generated artifacts.
  • [COMMAND_EXECUTION]: The skill uses git status --short to verify the state of generated files, which is a benign and standard development operation.
  • [SAFE]: The skill generates self-contained HTML walkthroughs that are kept local by default, avoiding data exfiltration and external dependencies like CDNs or remote fonts.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 03:38 PM
Security Audit — agent-trust-hub — pr-visual-explainer