Skills
Navigate
skills/google-labs-code/stitch-skills/stitch::extract-design-md/Gen Agent Trust Hub

stitch::extract-design-md

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates the static analysis of local source code to document visual design patterns. It operates on local files and does not execute the source code it analyzes.
  • [COMMAND_EXECUTION]: The Bash tool is intended for project discovery, such as identifying the framework via package.json or mapping directory structures. No dangerous command patterns or argument injections were identified.
  • [PROMPT_INJECTION]: The instructions do not contain any patterns designed to override agent safety protocols or bypass constraints. The metadata and instructional text are focused on the stated design extraction purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 05:39 PM
Security Audit — agent-trust-hub — stitch::extract-design-md