ai-podcast-creation
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references an external installation script for the
infshCLI hosted on GitHub. This is a neutral reference to official documentation from the service provider. - [COMMAND_EXECUTION]: The skill uses the
infshCLI through theBashtool to run audio generation and editing applications. This is expected behavior for its stated purpose. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface when processing user-provided documents for podcast script generation.
- Ingestion points: The prompt in the 'NotebookLM-Style Content' section accepts a
<your-document-content>variable which is interpolated directly into the model input. - Boundary markers: The example uses a simple 'Document:' prefix without robust delimiters or instructions to ignore embedded instructions.
- Capability inventory: The skill uses the
infsh app runcommand which executes applications based on generated scripts, creating a potential path for malformed instructions to affect tool parameters. - Sanitization: No input sanitization, character escaping, or validation logic is defined in the prompt template.
Audit Metadata