Build AI chat interfaces using ai-elements components — conversations, messages, tool displays, prompt inputs, and more. Use when the user wants to build a chatbot, AI assistant UI, or any AI-powered chat interface.
中文简介正在准备中。当前显示原文说明。
Interview the user relentlessly about a plan or design until reaching shared understanding. Use when fleshing out an idea, stress-testing a design, or when the user says "grill me".
中文简介正在准备中。当前显示原文说明。
查看安装方法安装前可以查看交给 AI 代理的指令内容。
Walk the design tree branch by branch. Resolve dependencies between intents/desires/decisions, one by one. Be Socratic — question premises, not just requirements. The user's why matters as much or more than their what: knowing motivation lets you suggest alternatives they haven't considered.
For each question, provide your own recommended answer based on what you know. If the codebase can answer the question, explore it instead of asking.
Actively surface constraints, limitations, and irreversible choices — these are strategic inputs, not just risk factors. They shape what's possible, which shapes what's worth doing.
Name anti-patterns when you see a design drifting toward one. God objects, feature envy, leaky abstractions, premature optimization, stringly-typed interfaces — name the failure mode, explain why it fails here, probe for agreement. Elimination sharpens the solution space faster than enumeration.
Sharpen the lexicon as you go. When a concept surfaces, pin it to a canonical term — propose one, test the user's reaction, note resistance. Vague language hides vague thinking. If memory/SPEC.md exists, pressure-test its §Lexicon against what the conversation reveals.
Do not create planning artifacts here. Durable conclusions from grilling promote into memory/SPEC.md or memory/PLAN.md through the next routed skill.
Pure elicitation — do not produce a plan or document.
When understanding is reached, present these options to the user (use tool-ask-question):
| # | Label | Target | Why |
|---|---|---|---|
| 1 | Write a spec | ln-spec | Understanding is sufficient for a spec |
| 2 | Plan frontier | ln-plan | Problem is clear, needs frontier breakdown |
| 3 | Scope one slice | ln-scope | One slice is already obvious |
Recommended: choose based on how much structure the understanding needs.
Adapted from mattpocock/skills/grill-me.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Build AI chat interfaces using ai-elements components — conversations, messages, tool displays, prompt inputs, and more. Use when the user wants to build a chatbot, AI assistant UI, or any AI-powered chat interface.
中文简介正在准备中。当前显示原文说明。
Search the live web via Perplexity Search API. Use when you need current documentation, release notes, vendor pages, news, domain-constrained web search, or date/recency filtering. Not for local codebase search or stable docs already in context.
中文简介正在准备中。当前显示原文说明。
Chrome DevTools CLI for browser automation via shell commands. Use when interacting with web pages from the command line — navigating, clicking, filling forms, inspecting console/network, taking screenshots, or extracting page content. Triggers on: browse a page, automate Chrome, inspect console, check network requests, take a screenshot, fill a form, click a button.
中文简介正在准备中。当前显示原文说明。
Uses the chrome-devtools-axi CLI for browser automation, accessibility-tree snapshots, console and network inspection, screenshots, Lighthouse audits, and performance traces. Use when interacting with Chrome from the shell, especially when the user mentions chrome-devtools-axi, AX snapshots, browser debugging, or DevTools automation from the command line.
中文简介正在准备中。当前显示原文说明。
Deep expertise in cmux — the terminal multiplexer with native browser views. Use when managing panes, reading terminal output, sending keystrokes, opening browser views, or manually testing web UIs and TUIs inside cmux. Triggers on: cmux, open a browser pane, split terminal, read screen, send keys, test this UI in cmux, preview in cmux.
中文简介正在准备中。当前显示原文说明。
Uses the gh-axi CLI for GitHub shell operations: issue, pull request, workflow run, release, repo, search, and API tasks. Prefer this over regular `gh` for GitHub reads and simple mutations when an agent needs compact, structured, suggestion-rich output. Triggers on: gh, GitHub CLI, github issue, github pr, pull request, workflow run, github release, gh api, repo inspection, list PRs, view issue, check workflow runs, inspect repo, GitHub shell operations.
中文简介正在准备中。当前显示原文说明。