consuming-endpoints-from-client-code

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates processing external data (OpenAPI specifications via endpoint-openapi-spec) which is then used for client code generation, creating a theoretical indirect injection surface. The risk is assessed as safe given the intended developer workflow and trusted source context. Ingestion points: endpoint-openapi-spec results are used for code generation; Boundary markers: None specified; Capability inventory: endpoint-run (network operations) and file management; Sanitization: Relies on third-party generators.
  • [SAFE]: The documentation emphasizes security best practices by explicitly advising users not to include personal API keys in client-side code and recommending server-side routing.
  • [SAFE]: The recommended third-party tools for client generation (e.g., openapi-generator-cli, @hey-api/openapi-ts) are standard, well-known developer utilities.
  • [SAFE]: The skill's external references (posthog.com) and tools are consistent with the infrastructure of the author (posthog).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 03:59 PM
Security Audit — agent-trust-hub — consuming-endpoints-from-client-code