prompt-refinery
Intercept and improve any user prompt before execution — rewriting it to be more durable, strategic, context-aware, and efficient.
When this skill is meant to trigger
Intercept and improve any user prompt before execution — rewriting it to be more durable, strategic, context-aware, and efficient. Acts as a prompt engineering harness that transforms weak or vague requests into precise, high-leverage instructions that elicit Claude's best work. Trigger on: 'improve this prompt', 'refine my prompt', 'make this prompt better', 'prompt refinery', 'harness this', 'optimize my request', 'sharpen this prompt', 'prompt engineering', 'rewrite my prompt', 'enhance this prompt', 'before you do this, improve it', 'prompt audit', 'strengthen this request', 'make this more effective'. Also trigger when a user explicitly pastes a prompt and asks for it to be improved, or says things like 'can you make this ask better', 'how would you rewrite this', 'clean up this instruction', or 'level up this prompt'. Use proactively when the user's request would clearly benefit from restructuring before execution.
How this skill works
Takes a draft prompt and returns a tighter, clearer version with the same intent. Expect an improved prompt you can copy, plus brief notes on what changed. Use the paste field for the prompt to refine.
Runs end to end from what you paste here.
Fast and cheap. The default for most skills. Catalog rates $0.3 in and $2.5 out per million tokens. The receipt uses OpenRouter's billed cost.
Paste your input and run the skill. The result is written here as it streams.