The Agent Skills Directory

[SAFE] (SAFE): The skill is purely instructional and template-based. It does not contain obfuscated code, unauthorized network requests, or credential harvesting.
[Indirect Prompt Injection] (LOW): The skill defines a command /generate-tests <file-path> and requests Read, Write, and Edit permissions. While it primarily provides templates, the ingestion of user-provided file paths is a standard entry point for indirect prompt injection if the processing agent does not sanitize input. However, the skill's content is focused on legitimate software development practices.
[COMMAND_EXECUTION] (SAFE): The skill lists commands for running tests (e.g., pytest, jest). These are standard developer tools and are presented as usage examples rather than hidden or malicious execution patterns.
[EXTERNAL_DOWNLOADS] (SAFE): No automated downloads or installation of third-party packages from untrusted sources are performed. References to pytest and Jest are standard in the development ecosystem.

test-generator