The Agent Skills Directory

[Command Execution]: The skill uses specialized sub-agents like code-testing-builder and code-testing-tester to compile and run generated tests. This is a core part of its functionality to ensure the generated code is functional and correct.
[Code Generation and Execution]: The pipeline automatically writes new test files and subsequently executes them. While this involves dynamic execution, the skill's instructions emphasize unit testing with mocked dependencies and the avoidance of external network calls to mitigate risks associated with executing generated code.
[Processing Untrusted Data]: The skill ingests source code from the local environment to generate unit tests. This process involves analyzing external input to determine test logic and structure. * Ingestion points: Source files analyzed during the Research and Implementation phases. * Boundary markers: The skill does not explicitly define delimiters for external content in its orchestration logic. * Capability inventory: The skill utilizes builder and tester sub-agents to compile and execute code. * Sanitization: No explicit sanitization or filtering of codebase content is mentioned before processing.

code-testing-agent