The Agent Skills Directory

[COMMAND_EXECUTION]: The workflow utilizes standard shell utilities (find, grep) to map the relationship between source files and their corresponding test files. These operations are restricted to the local src directory and are a standard part of the coverage analysis process.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze local source code to identify testing gaps. The instructions include specific checks for the agent to consider failure modes and malicious user behavior when generating tests, which acts as a built-in safety consideration.
Ingestion points: Reads the content of .ts and .tsx files located in the src directory.
Boundary markers: None explicitly provided; the agent relies on its internal logic to separate code analysis from instruction following.
Capability inventory: Local file system discovery via find and the generation of new test files.
Sanitization: No specific input sanitization is implemented for the source code being analyzed, as the agent is expected to treat the code as a subject for analysis rather than a source of instructions.
[DYNAMIC_EXECUTION]: The skill instructs the agent to generate new test scripts based on identified coverage gaps and predefined AAA (Arrange-Act-Assert) patterns. This code generation is the primary purpose of the skill and follows standard development practices for unit testing.

test-gaps