ai-podcast-creation
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references a command-line tool (
belt) and provides links to its official installation instructions on GitHub (https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md). These are well-known resources for the platform being utilized. - [COMMAND_EXECUTION]: The skill provides examples of running the
beltCLI tool to interact with various AI models (Kokoro TTS, AI music generation, Claude). These commands are standard for the skill's purpose and do not perform any hidden or dangerous operations. - [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface where external documents are processed to generate podcast scripts (e.g., using
openrouter/claude-sonnet-45). However, it provides standard interpolation patterns and does not introduce specific vulnerabilities beyond the inherent risks of processing untrusted text with an LLM.
Audit Metadata