youtube-full
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses a centralized API service (TranscriptAPI) for data retrieval, which is a standard and generally safer practice for cloud-based AI agents compared to local scraping.\n- [SAFE]: Credential management is handled through the agent's dedicated store, avoiding hardcoded secrets within the skill instructions.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources (YouTube transcripts, titles, and descriptions), creating an attack surface where malicious instructions could influence the agent's behavior.\n
- Ingestion points: Data entering the agent via
get_transcript,search_youtube, and channel-related operations inSKILL.md.\n - Boundary markers: The skill documentation lacks explicit instructions for the agent to use delimiters or ignore embedded instructions when processing the fetched content.\n
- Capability inventory: All tool operations across the skill focus on data retrieval; no dangerous capabilities such as file-system modification or arbitrary command execution are provided.\n
- Sanitization: No content validation or sanitization mechanisms are described for the data fetched from YouTube.
Audit Metadata