Podcast RSS Feed Transcriber
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches audio enclosures and metadata from external RSS feed URLs. This activity is a core requirement of the stated transcription service and targets user-provided or vendor-specified feeds.
- [PROMPT_INJECTION]: The skill processes untrusted external data (RSS metadata and transcribed audio), creating a surface for indirect prompt injection. 1. Ingestion points: Podcast RSS feeds and audio enclosures. 2. Boundary markers: The documentation does not specify the use of delimiters or 'ignore' instructions for external content. 3. Capability inventory: Includes network access for downloads and file system access for writing transcripts and maintaining a local SQLite database. 4. Sanitization: No explicit sanitization or validation methods are described for the processed text.
- [SAFE]: The skill metadata includes a self-claimed 'security_reviewed' status. This is evaluated as a non-authoritative claim provided by the skill author and does not influence the final security verdict.
Audit Metadata