The Agent Skills Directory

[SAFE]: The skill consists of markdown instructions and reference guides for software development practices. No executable code or suspicious patterns were detected.
[COMMAND_EXECUTION]: The skill instructs the agent to run testing tools such as pytest as part of the TDD cycle. This is the intended behavior for a development-focused skill and is considered safe within the context of coding assistance.
[PROMPT_INJECTION]: As a coding assistant skill, it naturally ingests user-provided code and test outputs, creating a surface for indirect prompt injection. This risk is inherent to coding tasks. * Ingestion points: Reads user-provided project files and test outputs. * Boundary markers: Not explicitly defined. * Capability inventory: Performs file writes and shell execution. * Sanitization: Not explicitly implemented.

dev-tdd