add-tts-engine
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the cloning of external code repositories from user-provided URLs to integrate new models.
- [COMMAND_EXECUTION]: The skill requires the execution of multiple shell commands for directory management, repository cloning, file inspection, and running the development environment, specifically using commands like git clone, mkdir, cat, and just dev.
- [DYNAMIC_EXECUTION]: The skill involves runtime environment configuration for frozen builds, including PyInstaller bundling instructions and monkey-patching library functions such as torch.load to manage hardware compatibility across different devices.
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from external repositories and local documentation, presenting a surface for indirect instructions to influence the agent.
- Ingestion points: Reads local developer documentation and clones external Git repositories for model integration.
- Boundary markers: The instructions do not define explicit delimiters or boundary markers for the ingested external code or documentation.
- Capability inventory: The skill has capabilities to modify the backend and frontend source code, update build configurations, and execute shell commands.
- Sanitization: A mandatory 'Phase 0: Dependency audit' step is required by the instructions to manually inspect external source code and dependencies before any code is integrated.
Audit Metadata