The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script scripts/coverage-analyzer.py uses the __import__('datetime') function to generate timestamps for generated reports. This is a form of dynamic loading, but because it targets a hardcoded standard library module, it is considered a low-risk development practice.
[INDIRECT_PROMPT_INJECTION]: The skill provides tools that read and analyze external project files, creating a potential surface for indirect prompt injection attacks where malicious code in a processed file could attempt to influence the agent.
Ingestion points: scripts/test-generator.py reads source code files to extract function definitions, and scripts/coverage-analyzer.py reads both source files and JSON coverage data.
Boundary markers: The scripts do not implement specific boundary markers or instructions to isolate the data within the files from the logic used by the agent or tools.
Capability inventory: The skill has the ability to read and write files and execute common developer tools (such as pytest, npm, and linting tools) via the scripts/test-runner.sh script.
Sanitization: The tools perform standard file reading and JSON parsing without specific security-focused sanitization or filtering of the file contents.

testing-strategies