shareAI

创建MCP服务器，让Claude通过统一协议调用外部工具和读取数据。涵盖Python或TypeScript实现、服务连接、Claude配置与运行验证。

• 为Claude添加自定义工具
• 连接外部API
• 读取数据库和配置信息

从安全性、正确性、性能、可维护性和测试质量检查代码，整理问题位置、影响、修复建议、改进项与已有优点，并给出是否适合合并的判断。

• 合并前审查代码变更
• 检查缺陷和边界条件
• 排查漏洞及依赖问题

为业务、研究、运营和创作设计并构建AI智能体，梳理所需操作与知识，从小规模配置开始，按实际需要逐步加入计划管理和子智能体。

• 设计业务助手
• 构建研究用智能体
• 梳理必要操作与知识

支持提取PDF正文、检查页数和文档信息、合并或按页拆分文件，也可将Markdown、HTML或Python生成的内容制作为PDF，并按页处理大文件。

• 提取PDF正文
• 将多个PDF合并为一个文件
• 将PDF按页拆分

Give slides, documents, reports, and HTML landing pages consistent colors and fonts. Choose from ten preset themes using a visual showcase, or create a custom theme from a description and review it before application.

中文简介正在准备中。当前显示英语简介。

• Making colors and fonts consistent across slides
• Choosing a theme for documents or reports
• Styling an HTML landing page

Diagnose missing language features in Kode by checking LSP, the protocol connecting code tools to language servers. Update plugin configuration, confirm servers are recognized, and test an operation such as jumping to a definition.

中文简介正在准备中。当前显示英语简介。

• Diagnosing missing LSP results in Kode
• Adding a language server through a plugin
• Reviewing a plugin's language server configuration

Build MCP servers, which connect AI agents to external services, using TypeScript or Python. Plan API access, design understandable tools and structured responses, implement authentication and error handling, and check behavior with the intended client.

中文简介正在准备中。当前显示英语简介。

• Building an MCP server for an external service
• Designing tool names, inputs, and responses for agents
• Implementing an MCP server in TypeScript or Python

Guide an AI agent through feature development, bug fixes, and refactoring with clear requirements, explained design choices, and small, tested changes. Keep human decisions central while making progress and verification results visible.

中文简介正在准备中。当前显示英语简介。

• Adding features to an existing project
• Tracing a bug to its root cause
• Refactoring code with tests in place

Test local web applications with Python scripts using Playwright, a browser automation tool. Verify interface behavior, investigate problems, capture screenshots, and inspect browser logs, with a helper that manages one or more application servers.

中文简介正在准备中。当前显示英语简介。

• Verify buttons, links, and form behavior
• Investigate problems in dynamic web interfaces
• Capture screenshots of local applications

Turn specialized knowledge and repeatable workflows into an Agent Skill. Work through concrete usage examples, write SKILL.md, organize reusable scripts and resources, and validate the package before producing a distributable .skill file.

中文简介正在准备中。当前显示英语简介。

• Turn recurring work into a reusable skill
• Package internal knowledge and API instructions
• Revise an existing skill after real use

Create original posters and static artwork by first defining a visual philosophy, then expressing it through color, shapes, spacing, and minimal text. Receive the artwork as PNG or PDF alongside a Markdown design statement.

中文简介正在准备中。当前显示英语简介。

• Create a poster with minimal text
• Express a theme through abstract artwork
• Develop several pages with a shared visual direction

Configure Kode features by describing the changes you want. The skill inspects settings, adjusts status displays, response styles, and plugins, and verifies changes through settings checks and small practical operations.

中文简介正在准备中。当前显示英语简介。

• Setting up or fixing the Kode status display
• Enabling TypeScript language support through plugins
• Changing the response style for a project

Build working web pages, components, and application interfaces around a clear visual direction. Shape typography, color, spacing, and animation to suit the audience and purpose, producing cohesive designs instead of repetitive, generic layouts.

中文简介正在准备中。当前显示英语简介。

• Creating a landing page with a distinct visual direction
• Designing and building a dashboard
• Styling React components consistently

Draft internal communications such as progress reports, leadership updates, company newsletters, FAQs, and incident reports. The skill selects guidance for the communication type and follows its instructions for format, tone, and information gathering.

中文简介正在准备中。当前显示英语简介。

• Preparing team updates on progress, plans, and problems
• Writing a company newsletter
• Drafting answers to internal FAQs

Investigate unexpected tool blocks and repeated permission prompts in Kode. The skill checks approved tools, command restrictions, inherited permissions, and session records, then verifies behavior by repeating the blocked action while preserving required user approval.

中文简介正在准备中。当前显示英语简介。

• Investigating an unexpectedly blocked tool
• Diagnosing repeated permission prompts
• Comparing permissions for main agents and subagents

Review SKILL.md files and their supporting resources using eight quality dimensions. Get a score out of 120, evidence for each assessment, and prioritized suggestions to improve specialized knowledge, activation descriptions, organization, and practical instructions.

中文简介正在准备中。当前显示英语简介。

• Review a skill before sharing it
• Improve descriptions that fail to trigger skill use
• Find redundant explanations in SKILL.md

Turn scattered notes and discussions into proposals, requirements documents, or decision records. The workflow gathers essential context, builds an outline, drafts sections with feedback, and checks whether readers without prior knowledge can understand the result.

中文简介正在准备中。当前显示英语简介。

• Turning discussion notes into a team document
• Writing product requirements or technical proposals
• Recording decisions and tradeoffs

A sample skill for demonstration

中文简介正在准备中。当前显示原文说明。

Recover and review local human-AI conversations from Claude Code, Codex, opencode, Grok Build, and Cursor. Use when the user wants to revisit, search, extract, summarize, or compare work across Agent session stores, time ranges, projects, or named conversations. Follow the user's requested output; question clustering, problem-space mapping, artifact review, drift analysis, and insights are optional lenses.

中文简介正在准备中。当前显示原文说明。

Reconstruct and report long-running or multi-turn research, architecture questions, reviews, decisions, completion results, and status as a clear, self-contained brief. Use when a reader must re-enter earlier context, understand the real question and its relationships, distinguish verified facts from judgment and open evidence, or make a decision without reconstructing the work log. Quote the relevant original user wording before substantive analysis, keep one coherent main line, use spatial diagrams when relationships matter, and finish with an explicit decision boundary and practical next step.

中文简介正在准备中。当前显示原文说明。

Transform an AI agent into a disciplined software development partner with strong judgment, transparent decisions, proportionate verification, and craftsmanship. Use for software implementation, feature work, bug fixing, refactoring, optimization, migrations, and sustained collaboration inside a codebase. Do not use merely because a task involves an AI agent; use a dedicated agent-design skill when the primary problem is agent architecture.

中文简介正在准备中。当前显示原文说明。

Coach leaders and managers to prepare, run, and review high-bandwidth meetings that advance real work and decisions. Use for meeting retrospectives, employee work reviews, decision meetings, agenda and question design, direct feedback, meeting-quality diagnosis, or repeated low-value discussions. Preserve direct standards without taking over the worker's work or degrading the truth signal.

中文简介正在准备中。当前显示原文说明。

Research why neural architectures and training methods work through forward computation, geometry, gradients, optimization dynamics, historical experiments, and competing explanations. Use for deep mechanism questions about neural networks, AI models, layers, representations, or training; explain the meaning in plain language with evidence and explicit limits.

中文简介正在准备中。当前显示原文说明。

Deeply research technical architecture, source code, mechanisms, SDKs, frameworks, project comparisons, and system-design options across repositories, history, official docs, issues, discussions, and high-quality community signals. Use when the user asks for deep research, source-level understanding, architecture comparison, mechanism satisfaction, or evidence for designing a system. Always map the question and obtain confirmation before beginning substantive research; skip for simple factual lookups or straightforward implementation tasks.

中文简介正在准备中。当前显示原文说明。