tiktok-video-search
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill interacts with the Frevana API at ai-factory.frevana.com. This interaction is restricted to the official API contract and uses secure credential handling.
- [COMMAND_EXECUTION]: The test suite uses the subprocess module to run the script during integration testing. This usage is restricted to the test environment and follows safe argument-passing techniques.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external video metadata from TikTok. Although this is a standard surface for indirect prompt injection in search tools, the skill's restricted scope and limited toolset minimize the risk. Mandatory Evidence Chain: 1. Ingestion points: Video metadata retrieved in scripts/tiktok_video_search.py. 2. Boundary markers: None. 3. Capability inventory: Network requests and file writes. 4. Sanitization: Standard JSON parsing.
Audit Metadata