# AgentSkillClub full listing

Updated 2026-11-07.

## Academy guide

Recommends matching Claude Academy courses, tutorials, and use cases when someone asks how to use Claude.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/academy-guide/
- Source: https://github.com/anthropics/skills/blob/main/skills/academy-guide/SKILL.md
- Raw: https://agentskillclub.com/skills/academy-guide/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Algorithmic art

Creates original p5.js art with seeded randomness and interactive parameters, for flow fields, particle systems, and similar requests.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/algorithmic-art/
- Source: https://github.com/anthropics/skills/blob/main/skills/algorithmic-art/SKILL.md
- Raw: https://agentskillclub.com/skills/algorithmic-art/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Brand guidelines

Applies Anthropic's brand colors and typography when an artifact should look like Anthropic's own materials.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/brand-guidelines/
- Source: https://github.com/anthropics/skills/blob/main/skills/brand-guidelines/SKILL.md
- Raw: https://agentskillclub.com/skills/brand-guidelines/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Canvas design

Creates original posters and other static pieces as PNG or PDF. The skill tells the agent to make new work rather than copy an existing artist.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/canvas-design/
- Source: https://github.com/anthropics/skills/blob/main/skills/canvas-design/SKILL.md
- Raw: https://agentskillclub.com/skills/canvas-design/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Claude API

Reference for the Claude API and Anthropic SDK: model ids, parameters, streaming, tool use, MCP, agents, caching, and model migration.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/claude-api/
- Source: https://github.com/anthropics/skills/blob/main/skills/claude-api/SKILL.md
- Raw: https://agentskillclub.com/skills/claude-api/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Discernment nudge

After a substantive answer or draft, adds two or three short follow-up questions so a person can check facts, probe the reasoning, and notice missing context.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/discernment-nudge/
- Source: https://github.com/anthropics/skills/blob/main/skills/discernment-nudge/SKILL.md
- Raw: https://agentskillclub.com/skills/discernment-nudge/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Doc co-authoring

Walks through a structured workflow for co-authoring documentation, proposals, specs, and decision docs.

- Type: skill
- License: Unstated
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/doc-coauthoring/
- Source: https://github.com/anthropics/skills/blob/main/skills/doc-coauthoring/SKILL.md
- Caution: This folder has no LICENSE.txt. The repository README says many skills are Apache-2.0, which is not a license file for this directory. The raw SKILL.md is not mirrored here.

## Word documents

Creates, reads, and edits Word documents (.docx) and templates (.dotx), including professional formatting.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/docx/
- Source: https://github.com/anthropics/skills/blob/main/skills/docx/SKILL.md
- Raw: https://agentskillclub.com/skills/docx/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Frontend design

Pushes UI work toward a deliberate visual direction, typography, and choices that do not look like a default template.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/frontend-design/
- Source: https://github.com/anthropics/skills/blob/main/skills/frontend-design/SKILL.md
- Raw: https://agentskillclub.com/skills/frontend-design/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Internal comms

Drafts internal communications in a company's existing formats: status reports, leadership updates, FAQs, incident notes, and newsletters.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/internal-comms/
- Source: https://github.com/anthropics/skills/blob/main/skills/internal-comms/SKILL.md
- Raw: https://agentskillclub.com/skills/internal-comms/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## MCP builder

Guides the building of an MCP server in Python (FastMCP) or Node/TypeScript (MCP SDK), with an emphasis on tool design.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/mcp-builder/
- Source: https://github.com/anthropics/skills/blob/main/skills/mcp-builder/SKILL.md
- Raw: https://agentskillclub.com/skills/mcp-builder/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## PDF

Reads, creates, merges, splits, fills, and encrypts PDF files, including forms and extracted text or tables.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/pdf/
- Source: https://github.com/anthropics/skills/blob/main/skills/pdf/SKILL.md
- Raw: https://agentskillclub.com/skills/pdf/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## PowerPoint

Creates, reads, and edits PowerPoint decks (.pptx) and templates (.potx).

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/pptx/
- Source: https://github.com/anthropics/skills/blob/main/skills/pptx/SKILL.md
- Raw: https://agentskillclub.com/skills/pptx/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Skill creator

Creates, edits, and evaluates agent skills, including description tuning and simple benchmarks.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/skill-creator/
- Source: https://github.com/anthropics/skills/blob/main/skills/skill-creator/SKILL.md
- Raw: https://agentskillclub.com/skills/skill-creator/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Slack GIF creator

Builds animated GIFs that fit Slack's size and animation constraints, and checks the file before it is sent.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/slack-gif-creator/
- Source: https://github.com/anthropics/skills/blob/main/skills/slack-gif-creator/SKILL.md
- Raw: https://agentskillclub.com/skills/slack-gif-creator/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Theme factory

Applies one of ten preset themes, or a theme it generates, to slides, docs, reports, and HTML pages.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/theme-factory/
- Source: https://github.com/anthropics/skills/blob/main/skills/theme-factory/SKILL.md
- Raw: https://agentskillclub.com/skills/theme-factory/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Web artifacts builder

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

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/web-artifacts-builder/
- Source: https://github.com/anthropics/skills/blob/main/skills/web-artifacts-builder/SKILL.md
- Raw: https://agentskillclub.com/skills/web-artifacts-builder/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Web app testing

Drives a local web app with Playwright to check behavior, capture screenshots, and read browser logs.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/webapp-testing/
- Source: https://github.com/anthropics/skills/blob/main/skills/webapp-testing/SKILL.md
- Raw: https://agentskillclub.com/skills/webapp-testing/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Spreadsheets

Reads, edits, and creates spreadsheets (.xlsx, .xlsm, .csv, and .tsv), including formulas, formatting, and charts.

- Type: skill
- License: Apache-2.0
- Author: Anthropic
- Status: example
- Page: https://agentskillclub.com/skills/xlsx/
- Source: https://github.com/anthropics/skills/blob/main/skills/xlsx/SKILL.md
- Raw: https://agentskillclub.com/skills/xlsx/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## React composition patterns

React composition patterns for components that have grown too many boolean props.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/vercel-composition-patterns/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/composition-patterns/SKILL.md
- Raw: https://agentskillclub.com/skills/vercel-composition-patterns/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Deploy to Vercel

Deploys an app to Vercel, preferring a git push on a linked project and falling back to the Vercel CLI.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/deploy-to-vercel/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/deploy-to-vercel/SKILL.md
- Raw: https://agentskillclub.com/skills/deploy-to-vercel/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## React best practices

React and Next.js performance rules from Vercel Engineering, used when writing, reviewing, or refactoring React code.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/vercel-react-best-practices/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/react-best-practices/SKILL.md
- Raw: https://agentskillclub.com/skills/vercel-react-best-practices/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## React Native skills

React Native and Expo practices for lists, animation, and native modules.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/vercel-react-native-skills/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/react-native-skills/SKILL.md
- Raw: https://agentskillclub.com/skills/vercel-react-native-skills/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## React view transitions

Implements page and shared-element animation with React's View Transition API, including Next.js.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/vercel-react-view-transitions/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/react-view-transitions/SKILL.md
- Raw: https://agentskillclub.com/skills/vercel-react-view-transitions/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Vercel CLI with tokens

Uses the Vercel CLI with an access token for deploys and environment variables, without an interactive login.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/vercel-cli-with-tokens/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/vercel-cli-with-tokens/SKILL.md
- Raw: https://agentskillclub.com/skills/vercel-cli-with-tokens/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Vercel optimize

Audits a deployed Vercel project for cost and performance, starting from Vercel metrics and then the files those metrics point at.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/vercel-optimize/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/vercel-optimize/SKILL.md
- Raw: https://agentskillclub.com/skills/vercel-optimize/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Web design guidelines

Reviews UI code against web interface guidelines for accessibility, forms, focus, and performance.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/web-design-guidelines/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/web-design-guidelines/SKILL.md
- Raw: https://agentskillclub.com/skills/web-design-guidelines/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Writing guidelines

Reviews docs and prose against the Vercel writing handbook.

- Type: skill
- License: MIT
- Author: Vercel
- Status: official
- Page: https://agentskillclub.com/skills/writing-guidelines/
- Source: https://github.com/vercel-labs/agent-skills/blob/main/skills/writing-guidelines/SKILL.md
- Raw: https://agentskillclub.com/skills/writing-guidelines/SKILL.md
- Caution: Mirrored file is SKILL.md only. Scripts, fonts, and other assets stay in the source repo. Read the file before use, and do not run referenced scripts until a person has reviewed them.

## Everything

Reference MCP server that exercises prompts, resources, tools, and other protocol features. It is a client test server, not a product.

- Type: mcp
- License: MIT OR Apache-2.0
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/everything/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/everything
- Install: npx -y @modelcontextprotocol/server-everything
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready. Upstream describes it as a test server for MCP client builders, not a useful server for everyday work.

## Fetch

Reference server that fetches web pages and converts HTML to markdown for a model to read.

- Type: mcp
- License: MIT
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/fetch/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/fetch
- Install: uvx mcp-server-fetch
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready. The README warns that this server can access local and internal IP addresses and may expose sensitive data. It can also start Node.js on first use unless you turn that off.

## Filesystem

Reference server for reading and writing files inside directories you pass on the command line.

- Type: mcp
- License: MIT OR Apache-2.0
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/filesystem/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem
- Install: npx -y @modelcontextprotocol/server-filesystem /path/you/explicitly/allow
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready. Only pass directories you intend the agent to touch. The README documents --read-only and --confine-roots. Roots from older clients can widen access unless you confine them.

## Git

Reference server for reading, searching, and changing Git repositories from an agent.

- Type: mcp
- License: MIT
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/git/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/git
- Install: uvx mcp-server-git
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready. The README says the server is in early development and the tool list can change. It can stage, commit, and merge.

## Memory

Reference server that stores a local knowledge graph so a client can remember entities and observations across chats.

- Type: mcp
- License: MIT OR Apache-2.0
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/memory/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/memory
- Install: npx -y @modelcontextprotocol/server-memory
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready. The graph is stored on the machine running the server. Know which file that is before you put personal notes in it.

## Sequential Thinking

Reference server that gives a model a tool for revising a sequence of thoughts while it works through a problem.

- Type: mcp
- License: MIT OR Apache-2.0
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/sequential-thinking/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking
- Install: npx -y @modelcontextprotocol/server-sequential-thinking
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready.

## Time

Reference server for the current time and IANA timezone conversion.

- Type: mcp
- License: MIT
- Author: Model Context Protocol
- Status: reference
- Page: https://agentskillclub.com/mcp/time/
- Source: https://github.com/modelcontextprotocol/servers/tree/main/src/time
- Install: uvx mcp-server-time
- Caution: Reference implementation in modelcontextprotocol/servers. Upstream says these servers demonstrate the protocol and are not production-ready.

## GitHub

GitHub's official MCP server for repositories, issues, pull requests, Actions, and the rest of the platform you can already access.

- Type: mcp
- License: MIT
- Author: GitHub
- Status: official
- Page: https://agentskillclub.com/mcp/github/
- Source: https://github.com/github/github-mcp-server
- Install: https://api.githubcopilot.com/mcp/
- Caution: The remote server is hosted by GitHub and signs you in with OAuth or a personal access token. The README also documents a local server if your client cannot open a remote MCP URL. Grant the token only the access the task needs.

## Playwright

Microsoft's Playwright MCP server. The model drives a browser through the accessibility tree instead of a screenshot-only loop.

- Type: mcp
- License: Apache-2.0
- Author: Microsoft
- Status: official
- Page: https://agentskillclub.com/mcp/playwright/
- Source: https://github.com/microsoft/playwright-mcp
- Install: npx @playwright/mcp@latest
- Caution: The README notes that some coding agents are better served by the Playwright CLI than by this MCP server, because a CLI skill can spend fewer tokens. A browser server can click through sites the user is logged into.

## Cloudflare API

Official remote MCP server for the Cloudflare API. The README describes a code-mode server that covers the API with a small tool list.

- Type: mcp
- License: Apache-2.0
- Author: Cloudflare
- Status: official
- Page: https://agentskillclub.com/mcp/cloudflare/
- Source: https://github.com/cloudflare/mcp
- Install: https://mcp.cloudflare.com/mcp
- Caution: Connecting authorizes the server against a Cloudflare account. The README documents OAuth and an API-token option. Review the permissions the OAuth screen asks for.

## Hugging Face

Official Hugging Face MCP server for the Hub and Gradio applications.

- Type: mcp
- License: MIT
- Author: Hugging Face
- Status: official
- Page: https://agentskillclub.com/mcp/huggingface/
- Source: https://github.com/huggingface/hf-mcp-server
- Install: https://huggingface.co/mcp
- Caution: The README documents a remote URL (https://huggingface.co/mcp) and a local stdio package, npx @llmindset/hf-mcp-server. Confirm which one your client is starting.

## Brave Search

Official Brave Search MCP server for web, local, image, video, news, and related search.

- Type: mcp
- License: MIT
- Author: Brave Software
- Status: official
- Page: https://agentskillclub.com/mcp/brave-search/
- Source: https://github.com/brave/brave-search-mcp-server
- Install: npx -y @brave/brave-search-mcp-server --transport stdio
- Caution: Needs a Brave Search API key. The README's default transport is stdio; HTTP is opt-in. This replaced the archived Brave example that used to live in the reference servers repo.

## Grafana

Official Grafana MCP server for a Grafana instance and the ecosystem around it.

- Type: mcp
- License: Apache-2.0
- Author: Grafana
- Status: official
- Page: https://agentskillclub.com/mcp/grafana/
- Source: https://github.com/grafana/mcp-grafana
- Install: uvx mcp-grafana
- Caution: The server talks to your Grafana. Point it at the instance and credentials the README asks for, not at a production stack you have not scoped down.

## Neon

Official Neon MCP server for managing Lakebase Postgres projects, branches, and queries in language.

- Type: mcp
- License: MIT
- Author: Neon
- Status: official
- Page: https://agentskillclub.com/mcp/neon/
- Source: https://github.com/neondatabase/mcp-server-neon
- Install: https://mcp.neon.tech/mcp
- Caution: The README shows a hosted URL and category query parameters that widen what the server can do. Start from the smallest category set that does the job.

## Supabase

Official Supabase MCP server so an assistant can work against a Supabase project.

- Type: mcp
- License: Apache-2.0
- Author: Supabase
- Status: official
- Page: https://agentskillclub.com/mcp/supabase/
- Source: https://github.com/supabase/mcp
- Install: https://mcp.supabase.com/mcp
- Caution: The README tells you to read Supabase's security notes before connecting. It documents read_only and a project_ref query parameter. Do not connect an assistant to a production database with write tools on by default.

## Chrome DevTools

Official Chrome DevTools MCP server and CLI so a coding agent can inspect and drive a live Chrome browser.

- Type: mcp
- License: Apache-2.0
- Author: Google Chrome
- Status: official
- Page: https://agentskillclub.com/mcp/chrome-devtools/
- Source: https://github.com/ChromeDevTools/chrome-devtools-mcp
- Install: npm i chrome-devtools-mcp@latest -g
- Caution: A DevTools session can see pages, network traffic, and logged-in state in the browser profile it attaches to. Use a profile you are willing to expose to the agent.

## AWS MCP servers

AWS Labs open-source suite of MCP servers for AWS. The README also points at the Agent Toolkit for AWS as the successor to these servers, plugins, and skills.

- Type: mcp
- License: Apache-2.0
- Author: Amazon Web Services
- Status: official
- Page: https://agentskillclub.com/mcp/aws/
- Source: https://github.com/awslabs/mcp
- Caution: Do not run the repository root as if it were a single server. Pick a server from the README. The same README says the Agent Toolkit for AWS is now the successor. Check which one you mean before you grant AWS credentials.

## gcloud

Official gcloud MCP server so an assistant can work with a Google Cloud environment through the gcloud CLI.

- Type: mcp
- License: Apache-2.0
- Author: Google Cloud
- Status: official
- Page: https://agentskillclub.com/mcp/gcloud/
- Source: https://github.com/googleapis/gcloud-mcp
- Install: npx -y @google-cloud/gcloud-mcp
- Caution: The server uses the gcloud credentials on the machine. The README also documents npx @google-cloud/gcloud-mcp init for Gemini CLI. The active account is whatever gcloud is already using.

## Microsoft MCP servers

Official Microsoft MCP monorepo. It currently documents the Azure MCP server and the Microsoft Fabric MCP server.

- Type: mcp
- License: MIT
- Author: Microsoft
- Status: official
- Page: https://agentskillclub.com/mcp/microsoft/
- Source: https://github.com/microsoft/mcp
- Caution: This is a catalog of servers, not one binary. Start from the Azure or Fabric README inside the repo. Azure/azure-mcp was archived on 25 August 2025.

## MongoDB

Official MongoDB MCP server for MongoDB databases and Atlas.

- Type: mcp
- License: Apache-2.0
- Author: MongoDB
- Status: official
- Page: https://agentskillclub.com/mcp/mongodb/
- Source: https://github.com/mongodb-js/mongodb-mcp-server
- Install: npx -y mongodb-mcp-server@latest setup
- Caution: The README also documents a hosted Atlas connection that does not run locally. A local server with a connection string can read and change data. Use a user with the access the task needs.

## Elasticsearch

Elastic's MCP server for querying Elasticsearch. Upstream marks it deprecated.

- Type: mcp
- License: Apache-2.0
- Author: Elastic
- Status: deprecated
- Page: https://agentskillclub.com/mcp/elasticsearch/
- Source: https://github.com/elastic/mcp-server-elasticsearch
- Caution: The README says this server is deprecated and will only receive critical security updates. It points to the Elastic Agent Builder MCP endpoint in Elastic 9.2.0+ and Elasticsearch Serverless. Listed here so the deprecation is visible, not as a recommendation to start a new install.

## Docker MCP Gateway

Docker's MCP CLI plugin and gateway, used by Docker Desktop's MCP Toolkit to run catalog servers in containers.

- Type: mcp
- License: MIT
- Author: Docker
- Status: official
- Page: https://agentskillclub.com/mcp/docker/
- Source: https://github.com/docker/mcp-gateway
- Caution: The gateway runs other MCP servers. A container is not a full review of the server inside it. Read the catalog entry for whatever the gateway launches.

## Atlassian

Official remote MCP server for Jira, Confluence, and the rest of the Atlassian products the README lists.

- Type: mcp
- License: Apache-2.0
- Author: Atlassian
- Status: official
- Page: https://agentskillclub.com/mcp/atlassian/
- Source: https://github.com/atlassian/atlassian-mcp-server
- Install: https://mcp.atlassian.com/v2/mcp
- Caution: The README says v2 is the version to use for new setups. The legacy SSE endpoint was scheduled to end support after 30 June 2026. OAuth grants access to the Atlassian site you sign into.

## Pinecone

Official Pinecone developer MCP server for docs, index configuration, and code informed by that configuration.

- Type: mcp
- License: Apache-2.0
- Author: Pinecone
- Status: official
- Page: https://agentskillclub.com/mcp/pinecone/
- Source: https://github.com/pinecone-io/pinecone-mcp
- Install: npx -y @pinecone-database/mcp
- Caution: The README's manual run example passes PINECONE_API_KEY. That key reaches your Pinecone project.

## Chroma

Official Chroma MCP server for collections, vector search, and full-text search against a Chroma client you choose.

- Type: mcp
- License: Apache-2.0
- Author: Chroma
- Status: official
- Page: https://agentskillclub.com/mcp/chroma/
- Source: https://github.com/chroma-core/chroma-mcp
- Install: uvx chroma-mcp
- Caution: The README's first example is an ephemeral in-memory client. Persistent, HTTP, and Chroma Cloud modes are separate argument sets. The one-line command above does not connect to a cloud database.

## Notion

Notion's open-source self-hosted MCP server. The README says it is no longer maintained and points people at the hosted server.

- Type: mcp
- License: MIT
- Author: Notion
- Status: unmaintained
- Page: https://agentskillclub.com/mcp/notion/
- Source: https://github.com/makenotion/notion-mcp-server
- Caution: Upstream says this repository is no longer actively maintained and recommends the hosted remote MCP documented at https://developers.notion.com/docs/mcp. Listed so the open-source server and its status are both visible.

## Context7

Upstash Context7 pulls version-specific library docs and code examples into a coding agent.

- Type: mcp
- License: MIT
- Author: Upstash
- Status: official
- Page: https://agentskillclub.com/mcp/context7/
- Source: https://github.com/upstash/context7
- Install: https://mcp.context7.com/mcp
- Caution: The README documents a hosted URL that takes an API key in an Authorization header, a setup command (npx ctx7 setup), and the local package @upstash/context7-mcp. The hosted server sends library names you ask about to Context7.

## Jina

Official Jina remote MCP server for the Reader, Search, Embeddings, and Reranker APIs.

- Type: mcp
- License: Apache-2.0
- Author: Jina AI
- Status: official
- Page: https://agentskillclub.com/mcp/jina/
- Source: https://github.com/jina-ai/MCP
- Install: https://mcp.jina.ai/v1
- Caution: The README warns that some clients cannot pass env vars, and shows people pasting a Jina API key into the client config. Prefer a client that keeps the key out of the file you commit.

## Apify

Official Apify MCP server. An agent can search Apify Store and run Actors for scraping and extraction.

- Type: mcp
- License: MIT
- Author: Apify
- Status: official
- Page: https://agentskillclub.com/mcp/apify/
- Source: https://github.com/apify/apify-mcp-server
- Install: https://mcp.apify.com
- Caution: Actors can browse the web and spend Apify credit. The README documents a hosted OAuth server and a local stdio package. Read an Actor's input and pricing before the agent runs it.

## Firecrawl

Official Firecrawl MCP server for search, scrape, and page interaction, returned as context for an agent.

- Type: mcp
- License: MIT
- Author: Firecrawl
- Status: official
- Page: https://agentskillclub.com/mcp/firecrawl/
- Source: https://github.com/firecrawl/firecrawl-mcp-server
- Install: https://mcp.firecrawl.dev/v2/mcp
- Caution: The README documents a hosted endpoint, an OAuth variant, and a keyless endpoint with a smaller tool list. Scraping a URL can hit sites you did not mean to automate. Respect the target site's terms.

## Exa

Official Exa MCP server for web search and crawling.

- Type: mcp
- License: MIT
- Author: Exa Labs
- Status: official
- Page: https://agentskillclub.com/mcp/exa/
- Source: https://github.com/exa-labs/exa-mcp-server
- Install: https://mcp.exa.ai/mcp
- Caution: Hosted search. The README says most clients only need that URL. An API key may still be required depending on the account; check the README if the connection is refused.

## Linux Terminal

Public-domain prompt “Linux Terminal” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: f
- Status: official
- Page: https://agentskillclub.com/prompts/linux-terminal/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/linux-terminal/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a linux terminal. I will type commands and you will reply with what the terminal should show. I want you to only reply with the terminal output inside one unique code block, and nothing else. do not write explanations. do not type commands unless I instruct you to do so. when i need to tell you something in english, i will do so by putting text inside curly brackets {like this}. my first command is pwd

## JavaScript Console

Public-domain prompt “JavaScript Console” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: omerimzali
- Status: official
- Page: https://agentskillclub.com/prompts/javascript-console/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/javascript-console/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a javascript console. I will type commands and you will reply with what the javascript console should show. I want you to only reply with the terminal output inside one unique code block, and nothing else. do not write explanations. do not type commands unless I instruct you to do so. when i need to tell you something in english, i will do so by putting text inside curly brackets {like this}. my first command is console.log("Hello World");

## Python Interpreter

Public-domain prompt “Python Interpreter” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: bowrax
- Status: official
- Page: https://agentskillclub.com/prompts/python-interpreter/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/python-interpreter/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act like a Python interpreter. I will give you Python code, and you will execute it. Do not provide any explanations. Do not respond with anything except the output of the code. The first code is: "print('hello world!')"

## SQL Terminal

Public-domain prompt “SQL Terminal” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: sinanerdinc
- Status: official
- Page: https://agentskillclub.com/prompts/sql-terminal/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/sql-terminal/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a SQL terminal in front of an example database. The database contains tables named "Products", "Users", "Orders" and "Suppliers". I will type queries and you will reply with what the terminal would show. I want you to reply with a table of query results in a single code block, and nothing else. Do not write explanations. Do not type commands unless I instruct you to do so. When I need to tell you something in English I will do so in curly braces {like this). My first command is 'SELECT TOP 10 * FROM Products ORDER BY Id DESC'

## UX/UI Developer

Public-domain prompt “UX/UI Developer” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: devisasari
- Status: official
- Page: https://agentskillclub.com/prompts/ux-ui-developer/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/ux-ui-developer/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a UX/UI developer. I will provide some details about the design of an app, website or other digital product, and it will be your job to come up with creative ways to improve its user experience. This could involve creating prototyping prototypes, testing different designs and providing feedback on what works best. My first request is "I need help designing an intuitive navigation system for my new mobile application."

## Tech Writer

Public-domain prompt “Tech Writer” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: lucagonzalez
- Status: official
- Page: https://agentskillclub.com/prompts/tech-writer/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/tech-writer/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a tech writer. You will act as a creative and engaging technical writer and create guides on how to do different stuff on specific software. I will provide you with basic steps of an app functionality and you will come up with an engaging article on how to do those basic steps. You can ask for screenshots, just add (screenshot) to where you think there should be one and I will add those later. These are the first basic steps of the app functionality: "1.Click on the download button depending on your platform 2.Install the file. 3.Double click to open the app"

## English Translator and Improver

Public-domain prompt “English Translator and Improver” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: f
- Status: official
- Page: https://agentskillclub.com/prompts/english-translator/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/english-translator/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as an English translator, spelling corrector and improver. I will speak to you in any language and you will detect the language, translate it and answer in the corrected and improved version of my text, in English. I want you to replace my simplified A0-level words and sentences with more beautiful and elegant, upper level English words and sentences. Keep the meaning same, but make them more literary. I want you to only reply the correction, the improvements and nothing else, do not write explanations. My first sentence is "istanbulu cok seviyom burada olmak cok guzel"

## Code Reviewer

Public-domain prompt “Code Reviewer” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: rajudandigam
- Status: official
- Page: https://agentskillclub.com/prompts/code-reviewer/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/code-reviewer/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a Code reviewer who is experienced developer in the given code language. I will provide you with the code block or methods or code file along with the code language name, and I would like you to review the code and share the feedback, suggestions and alternative recommended approaches. Please write explanations behind the feedback or suggestions or alternative approaches.

## Accessibility Auditor

Public-domain prompt “Accessibility Auditor” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: rajudandigam
- Status: official
- Page: https://agentskillclub.com/prompts/accessibility-auditor/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/accessibility-auditor/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as an Accessibility Auditor who is a web accessibility expert and experienced accessibility engineer. I will provide you with the website link. I would like you to review and check compliance with WCAG 2.2 and Section 508. Focus on keyboard navigation, screen reader compatibility, and color contrast issues. Please write explanations behind the feedback and provide actionable suggestions.

## Conventional Commit Message Generator

Public-domain prompt “Conventional Commit Message Generator” from the prompts.chat dataset (formerly Awesome ChatGPT Prompts).

- Type: prompt
- License: CC0-1.0
- Author: jeff-nasseri
- Status: official
- Page: https://agentskillclub.com/prompts/conventional-commit/
- Source: https://github.com/f/awesome-chatgpt-prompts/blob/main/prompts.csv
- Raw: https://agentskillclub.com/prompts/conventional-commit/PROMPT.md
- Caution: This is role text from a public prompt library, not a vetted operating procedure. Read it before you paste it into a system prompt. It does not grant permission to ignore a model's actual safety policies.

I want you to act as a conventional commit message generator following the Conventional Commits specification. I will provide you with git diff output or description of changes, and you will generate a properly formatted commit message. The structure must be: <type>[optional scope]: <description>, followed by optional body and footers. Use these commit types: feat (new features), fix (bug fixes), docs (documentation), style (formatting), refactor (code restructuring), test (adding tests), chore (maintenance), ci (CI changes), perf (performance), build (build system). Include scope in parentheses when relevant (e.g., feat(api):). For breaking changes, add ! after type/scope or include BREAKING CHANGE: footer. The description should be imperative mood, lowercase, no period. Body should explain what and why, not how. Include relevant footers like Refs: #123, Reviewed-by:, etc. (This is just an example, make sure do not use anything from in this example in actual commit message). The output should only contains commit message. Do not include markdown code blocks in output. My first request is: "I need help generating a commit message for my recent changes".

## Read the live web

A reading order for search and page fetch: start with the smallest tool that can see the public web, and widen only when the task needs it.

- Type: use-case
- License: CC0-1.0
- Author: AgentSkillClub
- Status: club-note
- Page: https://agentskillclub.com/use-cases/read-the-web/
- Caution: A club note is a reading order, not a product. Follow the license and the caution on each tool it links to.

## When this is the job

Someone asks an agent to open a URL, search, or come back with sources. Pick one server. Do not turn on every search tool in the client.

## A reasonable order

- **Fetch** is the reference server that turns a URL into markdown. Its README warns that it can also reach local and internal addresses. Use it on public URLs, in a lab, until you have decided that warning is acceptable.
- **Brave Search**, **Exa**, **Jina**, and **Firecrawl** are company servers for search and reading. Each README names an API key or a hosted URL. Brave replaced the old archived example in the reference-servers repo.
- Stop when you have the page. A second crawler rarely makes the answer better.

## What not to add in the same step

Do not also mount the filesystem server unless a person has named the directories. Reading the web and writing the disk are different permissions.

Includes: fetch, brave-search, exa, jina, firecrawl

## Make the office file

Anthropic's PDF, Word, PowerPoint, and spreadsheet skills, and the gap between the mirrored SKILL.md and the scripts that stay upstream.

- Type: use-case
- License: CC0-1.0
- Author: AgentSkillClub
- Status: club-note
- Page: https://agentskillclub.com/use-cases/office-documents/
- Caution: A club note is a reading order, not a product. Follow the license and the caution on each tool it links to.

## When this is the job

The output is a PDF, a Word document, a deck, or a spreadsheet. Anthropic publishes these four as Apache-2.0 skills in `anthropics/skills`, and this site mirrors each SKILL.md plus that directory's LICENSE.txt.

## Before you run one

The skill text calls scripts that are not in the mirror. Open the source folder and read those scripts. A skill that fills a PDF or builds a workbook is still code.

## How they differ

- **PDF** reads, merges, splits, fills forms, and creates PDFs.
- **Word documents** targets `.docx` and `.dotx`.
- **PowerPoint** targets `.pptx` and `.potx`.
- **Spreadsheets** targets `.xlsx`, `.xlsm`, `.csv`, and `.tsv`.

Theme factory and canvas design sit next to this set when the file also has to look like a designed artifact. They are separate skills.

Includes: pdf, docx, pptx, xlsx

## Build a UI and look at it

Design and React skills for the decisions, then a browser tool when someone needs to see the page running.

- Type: use-case
- License: CC0-1.0
- Author: AgentSkillClub
- Status: club-note
- Page: https://agentskillclub.com/use-cases/build-and-check-ui/
- Caution: A club note is a reading order, not a product. Follow the license and the caution on each tool it links to.

## When this is the job

An agent is about to invent a screen, review one, or prove that a click works.

## Decisions, then a browser

- **Frontend design** (Anthropic) is about visual direction so the page does not collapse into a generic template.
- **Web design guidelines** (Vercel) reviews accessibility, focus, forms, and similar interface rules.
- **React best practices** (Vercel) applies when the stack is React or Next.js and the question is performance.
- **Web app testing** is Anthropic's Playwright skill for local apps. **Playwright MCP** is Microsoft's server. The Playwright README notes that some coding agents do better with the Playwright CLI, because it spends fewer tokens than a standing MCP session.
- **Chrome DevTools** is the official Chrome server for a live browser. It can see the profile it attaches to.

Use one browser tool. Two of them fighting over the same page is how sessions get confused.

Includes: frontend-design, web-design-guidelines, vercel-react-best-practices, webapp-testing, playwright, chrome-devtools

## Work inside a repository

Filesystem, Git, and GitHub's official server, with deploy kept as a separate decision.

- Type: use-case
- License: CC0-1.0
- Author: AgentSkillClub
- Status: club-note
- Page: https://agentskillclub.com/use-cases/work-in-a-repo/
- Caution: A club note is a reading order, not a product. Follow the license and the caution on each tool it links to.

## When this is the job

The agent needs to read a checkout, make a commit, or open a pull request.

## Split the permissions

- **Filesystem** is a reference server. Pass the directories on the command line. The README documents `--read-only` when the agent should not write, and `--confine-roots` so an older client cannot widen the allow-list.
- **Git** is the reference server for status, diff, add, commit, and related operations. The README calls it early development. It can change the repository.
- **GitHub** is GitHub's own server, with a remote URL at `https://api.githubcopilot.com/mcp/` and a local option in the README. OAuth or a personal access token is the key. Scope it down.
- **Deploy to Vercel** is a skill for projects that are supposed to ship there. It is not part of reading a repo.

The filesystem and Git servers are reference implementations. Upstream says they are not production-ready.

Includes: filesystem, git, github, deploy-to-vercel

## Remember, and read current docs

Three different tools that all get called memory: a local graph, a vector store, and a live docs feed.

- Type: use-case
- License: CC0-1.0
- Author: AgentSkillClub
- Status: club-note
- Page: https://agentskillclub.com/use-cases/memory-and-docs/
- Caution: A club note is a reading order, not a product. Follow the license and the caution on each tool it links to.

## When this is the job

The agent is forgetting the repo between sessions, or it is inventing APIs from old training data. Those are different failures.

## Do not treat them as substitutes

- **Memory** is the reference knowledge-graph server. It stores entities and observations on the machine that runs it. It is not production-ready, and the file it writes is yours to locate.
- **Chroma** is Chroma's MCP server. `uvx chroma-mcp` is the README's ephemeral client. Persistent and cloud modes are different arguments.
- **Context7** is Upstash's docs server. It fetches version-specific library documentation. The hosted URL is `https://mcp.context7.com/mcp`. That is a network call, not a local notebook.

None of these replace reading the license of the code you are about to change.

Includes: memory, chroma, context7

## Add a skill or an MCP server

Anthropic's skill-creator and mcp-builder, plus the rule that a downloaded SKILL.md is text until someone reviews it.

- Type: use-case
- License: CC0-1.0
- Author: AgentSkillClub
- Status: club-note
- Page: https://agentskillclub.com/use-cases/make-a-skill/
- Caution: A club note is a reading order, not a product. Follow the license and the caution on each tool it links to.

## When this is the job

You want a new SKILL.md, or a new MCP server, instead of another ad-hoc prompt.

## Read these two first

- **Skill creator** is Anthropic's Apache-2.0 skill for writing, editing, and evaluating skills. The mirror on this site is the SKILL.md and the LICENSE.txt. The eval assets stay in the source repo.
- **MCP builder** is Anthropic's guide for a Python (FastMCP) or TypeScript MCP server. Same mirror rule.

## Then stop before you install it

This site's [install guide](/install/) is the sequence: fetch the JSON, read the license, download the raw file only when it is mirrored, and do not execute companion scripts until a person has read them. A skill that says "run this script" is still a script.

Includes: skill-creator, mcp-builder