OverviewHistoryStatsSecurity
Security
Independent checks from skills.sh's audit partners.
Independent checks from skills.sh's audit partners.
Warn3 providerslatest audit Aug 25, 2026
Independent checks from skills.sh's audit partners.
This skill provides a multi-step video annotation pipeline for generating training data. It involves processing videos through Vision-Language Models (VLMs) and Large Language Models (LLMs) to create captions, descriptions, and question-answer pairs. The primary security considerations involve the ingestion of untrusted data into prompts and the use of user-defined module paths for configuration.
Detected behaviors
No alerts
1 issue