youtube-edit-list
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of timestamped transcripts or notes. This creates a surface where malicious instructions embedded in the input could attempt to manipulate the agent's formatting or decision logic.
- Ingestion points: User-provided transcript or notes via the primary argument hint.
- Boundary markers: None identified in the skill body.
- Capability inventory: The skill generates markdown text output only and does not possess capabilities for file writing, network operations, or shell command execution.
- Sanitization: No input validation or filtering of transcript content is performed.
- [DATA_EXFILTRATION]: The skill instructions include reading from the local file path
~/.youtube-skills/metrics/<VIDEO_ID>.json. This access is scoped to the skill's own configuration and data directory (vendor resource) and is used for its primary function. No evidence of unauthorized access to system-level secrets or network exfiltration was detected.
Audit Metadata