The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses absolute and forceful language ('ABSOLUTELY MUST', 'NOT NEGOTIABLE', 'YOU DO NOT HAVE A CHOICE') to override the AI agent's standard operational guidelines and safety discretion.
[PROMPT_INJECTION]: It includes explicit behavioral overrides that command the agent to ignore its own rationalization process ('You cannot rationalize your way out of this') in favor of the specified workflow.
[COMMAND_EXECUTION]: The skill mandates the invocation of the 'Skill' tool to load external content based on a very low threshold ('1% chance' of relevance) before any other action, including clarifying questions. This requirement forces the agent into an architectural loop where external, unvetted instructions are given execution priority.

using-superpowers