The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill requires access to Bash(pytest:*) and Bash(npm:*) to execute the TDD cycle. This is a primary requirement for its functionality. While shell execution carries inherent risks, the scope is restricted to specific testing tools.
[Indirect Prompt Injection] (LOW):
Ingestion points: File names and test method names are used as arguments for shell commands (e.g., pytest tests/unit/test_user_service.py::test_create_user_returns_user -x).
Boundary markers: Absent. The skill does not instruct the agent to sanitize or delimit these inputs before execution.
Capability inventory: The skill has the ability to write/edit files and execute shell commands via pytest and npm.
Sanitization: Absent. If a test file name or test method were maliciously named to include shell metacharacters (e.g., ; curl...), it could lead to unintended command execution if the agent does not properly escape arguments.

tdd-workflow