claude-skills

Neural search via Exa MCP for web, code, and company research. Use when the user needs web search, code examples, company intel, people lookup, or AI-powered deep research with Exa's neural search engine.

Multi-source deep research using firecrawl and exa MCPs. Searches the web, synthesizes findings, and delivers cited reports with source attribution. Use when the user wants thorough research on any topic with evidence and citations.

搜索小红书 (Xiaohongshu) 笔记。**只要用户消息里出现「小红书」/「xhs」/「xiaohongshu」（任何形式、任何上下文）就必须触发本 skill 去搜**，不需要等 '搜/查/看看' 之类的动词。包括但不限于：'搜小红书 X' / '小红书搜 X' / '小红书查 X' / '小红书看看 X' / 'xhs 搜 X' / 'xhs search X' / '小红书上有没有 X' / '小红书怎么说' / 'xhs 上 X 怎么样' / '小红书有人说 X' / 用户随口提到 '小红书' 这三个字。从用户消息里推断查询词（去掉 '小红书'/'xhs' 这些触发词本身），然后调本地 Spider_XHS 项目搜出 title/author/likes/URL（必要时 desc/comments）。

Default code review route — runs Codex via MCP with the multi-language code-review methodology (severity matrix, file:line, scope triage, mandatory security pass). Read-only, no edits. Trigger when the user says "code review" / "审一下" / "review 一下" / "审这个 diff" / "用 codex 审" / "second opinion" / "复审" / "看看这次改动" or asks for any code review on a diff, file, or PR. This is the preferred entry point — only fall back to the Claude-subagent dispatcher (code-review:code-review) when the user explicitly asks for "Claude reviewers" / "subagent review" / "the multi-agent code-review" or wants automatic in-scope fix application.

Challenge an implementation approach or design choice via Codex MCP. Read-only, asks "is this the right path?" — not "are there bugs?". Trigger when the user says "质疑这个方案" / "这个设计靠不靠谱" / "challenge this design" / "design review" / "second-guess this approach" / "考虑过 X 吗" / "is this the right approach" / "stress-test the design" / "punch holes in this". Returns assumptions / failure modes / alternatives — NOT a bug list, NOT a severity matrix.

Triage issues through a state machine driven by triage roles. Use when user wants to create an issue, triage issues, review incoming bugs or feature requests, prepare issues for an AFK agent, or manage issue workflow.

Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.

Build a throwaway prototype to flush out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".

Turn the current conversation context into a PRD and publish it to the project issue tracker (or save as a markdown file). Use when user wants to create a PRD from the current context.

Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

UI/UX design intelligence. 67 styles, 96 palettes, 57 font pairings, 25 charts, 13 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient. Integrations: shadcn/ui MCP for component search and examples.

Git workflow patterns including branching strategies, commit conventions, merge vs rebase, conflict resolution, and collaborative development best practices for teams of all sizes.

Use this skill to measure performance baselines, detect regressions before/after PRs, and compare stack alternatives.

Analyze instincts and suggest or generate evolved structures

Export instincts from project/global scope to a file

Import instincts from file or URL into project/global scope

Show learned instincts (project + global) with confidence

Extract reusable patterns from the session, self-evaluate quality before saving, and determine the right save location (Global vs Project).

List known projects and their instinct statistics

Promote project-scoped instincts to global scope

Delete pending instincts older than 30 days that were never promoted

Analyze local git history to extract coding patterns and generate SKILL.md files. Local version of the Skill Creator GitHub App.

Show skill portfolio health dashboard with charts and analytics

Feishu (Lark) MCP integration: send messages, create groups, create documents, upload files, query users. Use when user mentions '飞书' or any Feishu/Lark operations.

Create dense, printable A4 landscape cheatsheets from Markdown using a custom syntax with multi-column layout. Use when the user asks to create a cheatsheet, revision sheet, crib sheet, or condensed reference document for exams, courses, or technical topics. Supports math formulas, tables, highlights, notes, and multi-page output. Converts Markdown to styled HTML and exports to PDF for printing.

Browser MCP routing guide: automatically select the right browser tool (Agent Browser for browsing/form-filling, Chrome DevTools for performance/debugging, Playwright for E2E testing) based on user intent. Use when user mentions browsing, performance analysis, debugging, or testing web pages.

Copilot Lint review workflow: check CI status → run local lint/build → review Copilot PR comments with critical thinking → fix only necessary issues → loop. Use when user says 'cl' or wants to review and fix Copilot code review comments.

Automated Git PR workflow: detect status → commit → create PR → wait for CI → fix failures → check Copilot comments → loop until all green. Use when user says 'cpr' or wants to push code through the full PR pipeline.

Agent Team implementation workflow for executing coding plans. Orchestrates Implementer, Reviewer, and Tester agents in sequence with worktree isolation. Use when: (1) user says 'impl', (2) user provides an implementation plan to execute, (3) user wants multi-agent coordinated development with review checkpoints.

AI-assisted writing workflow with four phases: Learn (Socratic dialogue → knowledge cards), Structure (article framework), Write (expand and polish), Finalize (assemble full article). Use when user says '写作', '写文章', '帮我写', 'vibe', or wants to co-author any written content.