snapshot-testing

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely educational, providing guidance on quality assurance techniques like characterization tests and visual regression. No executable scripts or dangerous commands were found in the skill files.
  • [EXTERNAL_DOWNLOADS]: The skill references several external documentation links in the SKILL.md frontmatter and the references/ file. These URLs target official documentation for well-known services and organizations, including Jest (Meta), Vitest, Playwright (Microsoft), Storybook, and Chromatic, which are considered safe sources.
  • [PROMPT_INJECTION]: The instructions do not contain any patterns attempting to override system prompts or bypass safety guidelines. The instructional language is focused on teaching testing disciplines and the 'approval cycle'.
  • [DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or network exfiltration patterns were identified. The network activity is limited to documentation references.
  • [DYNAMIC_EXECUTION]: There is no evidence of runtime code generation, compilation, or unsafe deserialization. The files consist of Markdown, YAML, and JSON data structures.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:35 AM
Security Audit — agent-trust-hub — snapshot-testing