ai-podcast
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to orchestrate media generation workflows through a specialized command-line interface (infsh app run). It executes commands for image training, text-to-speech synthesis, avatar animation, and video merging. All executed commands are strictly related to the documented podcast generation pipeline. - [EXTERNAL_DOWNLOADS]: The workflow processes external assets by passing URLs for images and audio files to various processing tools (e.g.,
phota/trainandpruna/p-video-avatar). These operations are intrinsic to the skill's functionality for processing user-supplied character references and generated media turns. - [SAFE]: The skill follows professional media production practices, including quality assurance steps, character consistency management, and resource optimization (sequential execution to avoid GPU OOM). It uses established services such as Google Gemini and Inworld for its core AI capabilities.
Audit Metadata