The Agent Skills Directory

[PROMPT_INJECTION]: The skill employs assertive and coercive language to bypass the agent's standard operating procedures and prioritize specific tool use. Evidence: The text uses terms like 'EXTREMELY-IMPORTANT', 'ABSOLUTELY MUST', 'not negotiable', and 'not optional' to enforce compliance. Evidence: Explicitly tells the agent 'You cannot rationalize your way out of this.'
[PROMPT_INJECTION]: Hijacks the agent's control flow by establishing an extremely low threshold for tool invocation, overriding the agent's autonomy. Evidence: Instructions state that if there is even a '1% chance' a skill might apply, the agent must invoke it before responding or taking any other action.
[PROMPT_INJECTION]: Explicitly instructs the agent to disregard internal reasoning and suppress standard conversational safety and clarification protocols. Evidence: The 'Red Flags' section categorizes logical thoughts (e.g., 'I need more context first' or 'This is just a simple question') as 'rationalizing' and instructs the agent to 'STOP'. Evidence: It mandates that 'Skill check comes BEFORE clarifying questions.'

using-superpowers