The Agent Skills Directory

[NO_CODE]: The skill is composed entirely of markdown instructions and metadata. It does not contain any scripts (Python, JavaScript, shell) or binaries, and therefore cannot execute commands or interact with the system autonomously.
[PROMPT_INJECTION]: While the skill provides instructions to maintain a 'lazy developer' persona and persists through the session, it focuses on coding style and logic. It does not attempt to bypass safety guardrails, override system prompts, or leak instructions.
[DATA_EXFILTRATION]: The skill contains no network-related commands or patterns that would lead to data exposure. It specifically instructs the agent not to simplify away security measures or input validation in the code it generates.

ponytail