api-designer
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides educational content and local utilities for API development. It includes several examples of API keys (e.g.,
sk_live_abcdef1234567890andsk_live_abc123def456) which are explicitly used as documentation placeholders and do not represent actual secrets. The Python scriptscripts/api_helper.pyuses the secureyaml.safe_loadmethod for processing specification files and performs only local file operations. - [INDIRECT_PROMPT_INJECTION]: The skill processes external API specifications (OpenAPI and GraphQL), which creates an inherent surface for indirect prompt injection if the processed data contains malicious instructions.
- Ingestion points: Specification files read by the
scripts/api_helper.pyscript and the agent. - Boundary markers: None present; the agent is not explicitly instructed to treat the content of the specifications as untrusted.
- Capability inventory: The skill provides local file read/write capabilities via the helper script.
- Sanitization: Data is parsed using standard JSON/YAML libraries, but no specific validation of the values within the schema is performed for instruction content.
Audit Metadata