ai-podcast-creation
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation includes instructions to install the 'belt' CLI from a well-known source and references configurations hosted on GitHub. These references point to established developer infrastructure for the purpose of tool installation.
- [COMMAND_EXECUTION]: The documentation demonstrates how to use the 'belt' command-line interface to interact with various AI models for text-to-speech, music generation, and audio editing. These commands are provided as examples for the user to follow in their own environment.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates workflows where LLMs (like Claude) generate podcast scripts from user-provided documents. While this represents a potential surface for indirect prompt injection, the skill acts as a bridge between specialized audio tools, and the instructions focus on standard content creation practices.
- Ingestion points: User-provided documents passed to LLM apps via the
beltCLI. - Boundary markers: The prompt templates use standard delimiters (e.g., 'Document:\n') to separate instructions from data.
- Capability inventory: The skill uses external AI tools for text-to-speech, music, and merging, which operate on the content generated by the LLM.
- Sanitization: Standard LLM prompt engineering is used to guide the generation process.
Audit Metadata