dialogue-audio
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill takes user-controlled text as input for the audio generation prompt, which is then processed by a downstream application.
- Ingestion points: The prompt field within the JSON input for the belt app run falai/dia-tts command.
- Boundary markers: None identified; user input is interpolated directly into the application's input structure.
- Capability inventory: The skill utilizes the belt CLI tool to execute audio generation and media merging tasks.
- Sanitization: There is no mention of sanitization, filtering, or validation for the text content provided by the user.
- [EXTERNAL_DOWNLOADS]: The skill fetches necessary components and documentation from external repositories associated with the service provider.
- Fetches installation instructions from a raw GitHub URL belonging to the inference-sh organization.
- Installs the belt CLI tool via npx from the belt-sh package scope.
- References related skills hosted within the inference-sh GitHub repository.
- [COMMAND_EXECUTION]: The skill requires the use of the belt command-line interface to interact with cloud-based AI services.
- Instructions include authentication via belt login.
- Core functionality is achieved through belt app run commands for speech synthesis and media processing.
- Service discovery is demonstrated using belt app list.
Audit Metadata