youtube-trending-scanner
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a shell command that includes user-provided input (the niche name) as a command-line argument. If the niche name is not properly sanitized or escaped by the agent, it could lead to arbitrary command execution on the host system.
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
google-api-python-clientpackage from the official Python Package Index (PyPI). This is a well-known library from a trusted organization used for its intended purpose. - [PROMPT_INJECTION]: The skill processes untrusted data from YouTube (video titles, tags, and channel names) and later instructs the agent to read this data to generate a report. This creates a surface for indirect prompt injection, where malicious content embedded in video metadata could influence the agent's behavior during the report generation phase.
- Ingestion points: Data is fetched from the YouTube Data API and saved to
trending_data.jsonin Step 7. - Boundary markers: None are specified for the ingested content in the report generation phase.
- Capability inventory: The skill performs shell executions (pip install, python execution, file removal) in Steps 4, 5, and 6.
- Sanitization: The Python script uses basic regex to sanitize the niche name for a directory path, but the shell command in Step 5 does not explicitly define sanitization for the niche argument.
Audit Metadata