Kỹ năng agent giúp AI trợ lý của bạn mạnh hơn

Adds short follow-up questions after substantive answers so users can check facts, assumptions, and missing context.

Accessible autonomous AI agents: build and run agents that pursue goals across multi-step workflows.

Open-source Python library for AI browser agents with stealth browsers and a hosted agent API.

AI-driven development platform: autonomous agents that write code, run commands, and open PRs.

Microsoft framework for agentic AI: orchestrate multi-agent conversations and workflows.

Framework for orchestrating role-playing autonomous AI agents that collaborate on complex tasks.

Guide for building high-quality MCP servers that let LLMs interact with external services through well-designed tools.

Official Model Context Protocol reference servers: filesystem, fetch, memory, and more.

Up-to-date documentation for AI agents: pull current library docs into any MCP client.

Chrome DevTools for coding agents: inspect, debug, and profile web apps from your AI assistant.

Official Playwright MCP server: browser automation and structured accessibility snapshots.

GitHub's official MCP Server: manage repos, issues, PRs, and workflows from AI assistants.

Provide Figma layout information to AI coding agents so generated UI matches designs.

Search, scrape, crawl, and map the web with JavaScript rendering from any MCP client.

Official Notion MCP Server: read, search, and update Notion pages and databases via AI.

Connect Supabase to your AI assistants: manage databases, schemas, and projects directly.

Agentic coding tool in your terminal: understands your codebase and handles tasks via natural language.

Open-source AI agent that brings the power of Gemini directly into your terminal.

AI pair programming in your terminal: edit code across files with automatic git commits.

Autonomous coding agent as an SDK, IDE extension, or CLI assistant for VS Code and JetBrains.

Open-source AI code editor: a Cursor alternative with inline agents and chat.

A whole dev team of AI agents inside your code editor, with custom modes and orchestration.

Reference skill for building with the Claude API: models, tool use, streaming, and integration patterns.

1000+ toolkits with tool search, authentication, and sandboxed execution for AI agents.

Test local web applications with Playwright: verify functionality, debug UI behavior, and capture screenshots.

Web testing and automation framework: test Chromium, Firefox, and WebKit with a single API.

Open-source PR reviewer: automated code review, suggestions, and analysis for pull requests.

Create, read, edit, and manipulate Word documents (.docx) including tracked changes, images, and find-and-replace.

Read, merge, split, rotate, watermark, fill, and OCR PDF files directly from your AI assistant.

Create, read, edit, and combine PowerPoint presentations and templates with speaker notes and layouts.

Create, edit, and fix spreadsheets (.xlsx, .xlsm, .csv) with formulas, charts, and clean formatting.

Web scraping framework handling everything from a single request to a full-scale crawl.

Let AI create and edit Excel files with charts, PivotTables, slicers, and formulas — no Excel required.

Open-source cloud security platform: automate security and compliance for AWS, Azure, and GCP.

Docker MCP CLI plugin and gateway: manage containers and images from AI assistants.

Structured workflow for co-authoring documentation, proposals, and technical specs with AI.

Write status reports, leadership updates, newsletters, FAQs, and incident reports in company formats.

Recommends matching courses and tutorials from Claude Academy when users ask how to use Claude products.

Fetch deepwiki.com knowledge for any GitHub repo directly in Cursor and other code editors.

Create new agent skills, improve existing ones, and run evals to benchmark triggering accuracy.

Guidance for distinctive, intentional UI design: aesthetic direction, typography, and non-templated choices.

Build elaborate multi-component claude.ai HTML artifacts with React, Tailwind CSS, and shadcn/ui.

Create posters, art, and static visual designs in .png and .pdf using design philosophy principles.

Generative art with p5.js: seeded randomness, flow fields, and interactive parameter exploration.

Style artifacts with 10 pre-set themes (colors and fonts) or generate a new theme on the fly.

Create animated GIFs optimized for Slack with constraints, validation tools, and animation concepts.

Apply Anthropic brand colors and typography to artifacts for a consistent company look-and-feel.