ecommerce-tiktok-video-search
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill runs a local Python script (
scripts/tiktok_video_search.py) to manage API requests and data handling. - [DATA_EXFILTRATION]: The skill transmits search queries to the EchoTik API proxy. This is the intended function of the tool and uses standard environment variables for authentication with the vendor's service.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external video data which could contain malicious instructions, though this is a standard risk for data-fetching tools.
- Ingestion points: Video descriptions and metadata retrieved from the API.
- Boundary markers: None; the agent processes the resulting JSON records.
- Capability inventory: File system writes to the nexscope directory and network access to the API proxy.
- Sanitization: The script does not perform content filtering on retrieved video data.
- [SAFE]: The skill includes well-defined local storage and caching logic that adheres to project structure conventions and respects credit limitations.
Audit Metadata