ai-podcast-creation
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external repositories for configuration and installation instructions.
- Fetches installation guidelines from
https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md. - Instructs the user to add the
belt-sh/clidependency using the platform's package manager. - [INDIRECT_PROMPT_INJECTION]: The skill describes workflows that ingest untrusted content, which could lead to unintended agent actions if the content contains malicious instructions.
- Ingestion points: The "NotebookLM-Style Content" workflow in
SKILL.mdprocesses external document content (<your-document-content>) to generate scripts. - Boundary markers: The workflow lacks explicit delimiters or instructions for the AI to ignore embedded commands within the provided document content.
- Capability inventory: The skill uses the
beltCLI tool to perform audio generation and writes the resulting data to local JSON files. - Sanitization: There is no evidence of sanitization or validation of the input document content before it is processed by the language model.
Audit Metadata