podcast-topic-landscape
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a helper script for API interactions that features strong defensive programming, such as using anchored regular expressions to whitelist allowed API routes and ensuring parameters are URL-encoded.
- [SAFE]: API credentials are handled securely by passing them to the network tool via a temporary configuration file with restricted permissions, which prevents exposure in process listings.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from external podcast platforms. Ingestion occurs via API calls in the helper script. The instructions include negative constraints to prevent the agent from fabricating information from this metadata. All input parameters are sanitized via URL-encoding.
Audit Metadata