The Agent Skills Directory

[PROMPT_INJECTION]: The skill focuses on standard software engineering practices. There are no instructions aimed at bypassing safety guardrails, overriding system-level instructions, or revealing underlying model prompts. The 'auto' mode provides the agent with autonomy to iterate through TDD cycles but does not grant permissions beyond the scope of the coding task.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied feature requests to generate code. It provides structured planning phases (ZOMBIES checklist) and implementation steps that act as natural reasoning boundaries. These steps minimize the risk of the agent misinterpreting user-provided code requirements as direct commands to the agent's core controller.
[EXTERNAL_DOWNLOADS]: The skill includes a markdown reference link to an external educational blog post by James Grenning regarding TDD principles. No executable code, scripts, or binary dependencies are downloaded or executed from external sources.
[COMMAND_EXECUTION]: The skill instructs the agent to run tests as part of the TDD workflow. This is standard behavior for a development-focused skill and is limited to the testing framework relevant to the user's project environment.
[SAFE]: All components of the skill, including the instructions and the local reference file, are consistent with its stated purpose of improving code quality through test-driven development.

tdd