stitch-sdk-development

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides documentation for a development workflow involving the @google/stitch-sdk vendor resource. It describes a generation pipeline that uses deterministic template expansion rather than dynamic execution or LLM-based generation.
  • [SAFE]: Secret management is handled securely by recommending environment variables (STITCH_API_KEY, STITCH_ACCESS_TOKEN) for authentication, avoiding hardcoded credentials.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes an architecture for processing external data from an MCP server. (1) Ingestion point: tools/list output captured in tools-manifest.json. (2) Boundary markers: stitch-sdk.lock file and validate-generated.ts consistency check. (3) Capability inventory: File system writes to packages/sdk/generated/ and network operations via StitchToolClient. (4) Sanitization: Input validation using Zod schemas.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 03:36 AM
Security Audit — agent-trust-hub — stitch-sdk-development