video-analysis
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill script
scripts/analyze_video.pyperforms network operations tohttps://geekai.co/api/v1/chat/completions(or a user-configured base URL) to access video analysis services. - [DATA_EXFILTRATION]: User-provided prompts and video URLs are transmitted to the external GeekAI API service. This behavior is the primary intended function of the skill for processing video content.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to its handling of untrusted external data that is subsequently passed to a large language model. ● Ingestion points: User-supplied prompts, video URLs, and data read from local JSON files (e.g., from
douyin-video-fetch) processed inscripts/analyze_video.py. ● Boundary markers: Input data is encapsulated within a structured JSON payload for the OpenAI-compatible API request. ● Capability inventory: The script usesurllib.requestto perform network POST requests to an external endpoint. ● Sanitization: No explicit validation or sanitization of input strings is performed before inclusion in the API request body.
Audit Metadata