The Agent Skills Directory

[PROMPT_INJECTION]: The skill employs extreme imperative language to override the agent's standard operational guidelines. Phrases like 'ABSOLUTELY MUST', 'not negotiable', and 'DO NOT HAVE A CHOICE' are used to force the agent to prioritize skill invocation over all other actions, including clarifying questions.
[PROMPT_INJECTION]: The 'Red Flags' section explicitly instructs the agent to ignore its own reasoning or 'rationalizing' thoughts (e.g., 'I need more context first', 'This is just a simple question'), which is a common tactic in prompt injection to bypass built-in safety filters or standard operating procedures.
[PROMPT_INJECTION]: By lowering the threshold for tool invocation to a '1% chance', the skill pressures the agent into executing external code or scripts before it has fully evaluated the user's intent or the safety of the task.

using-superpowers