design-context-extract

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: Analyzed for potential indirect prompt injection surfaces when processing external content.
  • Ingestion points: Fetches content from external URLs using WebFetch and scans local project files using Read and Glob.
  • Boundary markers: The skill does not define explicit delimiters to isolate external data from agent instructions.
  • Capability inventory: The agent has access to Bash, Write, and WebFetch tools.
  • Sanitization: No specific sanitization logic is defined for the fetched content.
  • [EXTERNAL_DOWNLOADS]: Mentions integration with Google Stitch, which is a service provided by a trusted organization.
  • [COMMAND_EXECUTION]: Includes the Bash tool in allowed-tools to support scanning project files and generating configuration files, which is appropriate for the skill's utility.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 11:41 PM