The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes standard testing commands (e.g., npx vitest, python -m pytest, go test, cargo test) based on the detected project environment. These operations are expected for a test-running utility.
[INDIRECT_PROMPT_INJECTION]: The skill reads local source code files to identify functions and components for test generation. This processes untrusted data (source code) which could theoretically contain instructions intended to influence the AI, but this is an inherent risk of code-analysis tools.
[DYNAMIC_EXECUTION]: The skill generates new test files and offers to execute them. This involves runtime creation and execution of code, which is the primary intended purpose of the test generation feature.

test-detect