ai-podcast-creation
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requests permission to execute shell commands via the
infshCLI. This is used to interact with the inference.sh platform for text-to-speech, music generation, and audio editing functions.- [EXTERNAL_DOWNLOADS]: The skill's documentation suggests installing additional tools and skills from the vendor's official repository (inference-sh/skills) using thenpxcommand.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection in the 'NotebookLM-Style Content' workflow where documents are converted into scripts. - Ingestion points: External document content is interpolated into the prompt for the
openrouter/claude-sonnet-45tool inSKILL.md. - Boundary markers: The prompt template lacks clear delimiters or protective instructions to separate the untrusted document content from the script-writing instructions.
- Capability inventory: The skill possesses the capability to execute shell commands (
infsh) and various AI processing tools. - Sanitization: No input validation or sanitization is performed on the provided document content before it is processed by the LLM.
Audit Metadata