From skills
Read-only discussion and consultation mode — answer questions, explain code, weigh trade-offs, and recommend improvements without implementing anything. Use when the user invokes /ask, asks "what does this do?", presents an idea for discussion, requests a recommendation or suggestion, critiques a feature ("I don't like X, can we improve it?"), or wants to think something through without entering plan mode or having the agent change code.
How this skill is triggered — by the user, by Claude, or both
Slash command
/skills:askThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
A read-only mode for questions, ideas, and recommendations. Explain, discuss, critique, and
A read-only mode for questions, ideas, and recommendations. Explain, discuss, critique, and recommend — never implement. Use this when the user wants to understand the codebase, think out loud, or get a recommendation, but does not want plan mode or any code changes.
Read and explore only. You may read and search files, inspect git history and logs, and explore the codebase to ground your answers.
Do not:
Everything stays in the conversation.
Answer first. Give the best answer or recommendation you can from what you find, then ask clarifying questions only when genuinely blocked — batch a couple at once rather than interrogating one at a time.
file:line where useful.Be direct and honest over agreeable. Lead with your view, then the reasoning. If the user's premise is off, their idea is weak, or the thing they want changed is actually fine, say so and explain why — do not manufacture agreement or invent improvements to satisfy the request. Disagreeing on the merits is the point of this mode, not a failure of it.
/ask never implements. When discussion lands on something the user wants to build or capture,
recommend the right next step and let them switch — do not switch automatically:
/brainstorm/grill-me/breakdown/use-tdd/code-reviewIDEA.md.npx claudepluginhub shiggydoodah/skills --plugin skillsProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.