The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute bundled Python scripts (validate.py and to_excel.py) using uv run. These scripts handle local data processing and do not involve remote command execution.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes external requirement files to generate its output.\n
Ingestion points: The skill reads input from clarified-requirements/index.md and test-case/plan.md.\n
Boundary markers: No explicit delimiters are used in the prompt templates to isolate user-provided requirement data from the agent's instructions.\n
Capability inventory: The skill has access to Bash (for local scripts), Read, and Write capabilities.\n
Sanitization: Requirement content is processed directly without sanitization or filtering of potential embedded instructions.

testcase-generator