Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From banthis
Use when user asks to ban, stop, or remember not to repeat an agent behavior ('stop doing X', 'you keep doing X', 'ban this'). Immediately capture it as a permanent negative rule using the banthis tool.
npx claudepluginhub agent-sh/banthis --plugin banthisHow this skill is triggered — by the user, by Claude, or both
Slash command
/banthis:banthis [behavior description][behavior description]This skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
When the user shows clear frustration with a recurring agent behavior, you must capture it using `banthis` so the rule persists across all future sessions.
Creates p5.js generative art with seeded randomness, noise fields, and interactive parameter exploration. Use for algorithmic art, flow fields, or particle systems.
Share bugs, ideas, or general feedback.
When the user shows clear frustration with a recurring agent behavior, you must capture it using banthis so the rule persists across all future sessions.
Use $ARGUMENTS as the behavior to ban. If it is empty, infer the behavior from the most recent explicit correction in the conversation.
$ARGUMENTS
No 'let me be honest' preambles)Do not X — reason.banthis if available, otherwise npx --yes github:agent-sh/banthis):banthis add "<title>" "<rule>"
Use --global only for behaviors that should apply to every project (verbal tics, generic LLM habits).
Banned: <title>banthis go into a managed section of CLAUDE.md or AGENTS.md and have higher priority than normal instructions.axiom).banthis init in projects if the meta-rule is not yet present (it teaches agents to invoke banthis automatically).axiom (positive knowledge capture)agnix (validates the resulting CLAUDE.md / AGENTS.md)skill-curator and system-prompt-curator (for improving how agents are instructed)