youtube-channel
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from YouTube (channel names, descriptions, and video titles), which could potentially contain malicious instructions.
- Ingestion points: Data retrieved from the YouTube API via the FlashData MCP endpoint in
SKILL.md. - Boundary markers: The skill does not implement explicit delimiters or boundary markers to separate retrieved content from the agent's instructions.
- Capability inventory: The skill maintains network capabilities via
httpxto communicate with the vendor's API. - Sanitization: The provided Python example does not perform sanitization or validation of the retrieved text before processing or display.
- [EXTERNAL_DOWNLOADS]: The skill documentation recommends installing standard Python packages and connecting to the vendor's hosted API service.
- Evidence: Instructions to install
mcpandhttpxfrom standard package registries. - Evidence: Establishes a connection to the vendor-owned infrastructure at
https://data.flashdata.dev/mcp.
Audit Metadata