innovationengines

grill-me

Interview the user with 10-15 targeted questions BEFORE building anything, then confirm a short spec, then build. Use this skill whenever the user asks Claude to build, create, make, code, design, draft, or generate anything non-trivial — an app, website, script, tool, document, presentation, campaign, plan, or system — even if the request looks detailed enough to start immediately. Especially trigger on vague one-line build requests like "make me a dashboard" or "build an app for my gym". Do NOT use for pure questions, explanations, debugging existing code, tiny edits, or reformatting.

how-to

Turn any 'I want to do X with Claude but I don't know how' into a finished result the user built with their own hands. Use this whenever someone names a goal they want to reach with Claude or AI and wants to be walked there until it is actually done. That covers building a Cowork workflow, automating a repetitive task, setting up a content or newsletter system, writing better prompts, connecting a tool, organizing files, or any 'how do I get Claude to' request. Trigger on phrases like 'how do I', 'I want to do this with Claude', 'give me the exact steps', 'walk me through', 'teach me to', 'I don't know how to', 'get me to the result', or the command /how-to. The skill maps the full step list first, then coaches one rookie-level step at a time and will not move on until the current step is finished and understood. Built for non-technical beginners.

prompt-master

Restructure messy, rambling, or stream-of-consciousness requests into a clean task spec BEFORE executing them, so nothing buried in the dump gets lost. Use this skill whenever the user's message is a brain dump: long unstructured paragraphs, voice-dictated text, multiple tangled tasks, requests with contradictions, "sorry this is messy but", walls of context with the actual ask buried in the middle, or copy-pasted notes ending in "can you do this". Do NOT use for short clear requests, follow-ups within an ongoing task, or casual conversation; restructuring a clean request is pure friction.

skill-audit

Review the user's installed Claude Skills for overlap, dead weight, vague triggers, and gaps, then deliver a keep/merge/fix/delete action plan. Use whenever the user types /skill-audit, says their skills overlap or fire at the wrong times, asks which skills to keep or delete, wants their skill library cleaned up or organized, or asks "which of my skills are actually useful". Also use after a user has installed a batch of new skills and wants to know how they fit together.

write-a-skill

Turn a plain-language description of a desired behavior into a properly structured, installable SKILL.md following Anthropic's skill-authoring best practices: trigger-optimized description, explain-the-why instructions, output templates, examples, and anti-trigger calibration. Use this skill whenever the user says "make a skill", "write a skill for", "turn this into a skill", "create a SKILL.md", describes a behavior they want Claude to follow every time, or complains about re-explaining the same instructions across chats (a recurring-behavior request is a skill request in disguise). Do NOT use for one-off task instructions, MCP server development, or general prompt-writing advice.