SureForge
SkillAn instruction-only skill that makes agents research before they ask, plan before they build, and verify before they call anything done.
- workflow
- verification
- quality
- skill
Full Chart
Filter by region or search by name, tagline, or tag.
331 resources
An instruction-only skill that makes agents research before they ask, plan before they build, and verify before they call anything done.
Wire all 15 Claude Code hook events into the Clawd on Desk desktop pet — and turn permission requests into a blocking desktop card.
An unofficial Rust 1:1 reimplementation of the Claude Code terminal UI, ported file-for-file onto an iocraft TUI framework.
Hook-level enforcement for the natural-language rules in AGENTS.md and CLAUDE.md — every edit is judged and violations are repaired in the same turn.
Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
An agent skill that geolocates a photo with nothing readable in it — OSM geometry, elevation skylines and satellite tiles turn “where was this taken?” into a measured answer with an error radius.
A Claude Code plugin that replaces the lossy compaction summary with Jev-scored pruning: stale tool calls are dropped, and everything kept stays verbatim.
A local launcher that points Claude Code and seven other coding harnesses at any OpenAI-, Anthropic- or Responses-compatible endpoint — without editing a single config file.
A skill that makes AI coding agents map the architecture and declare the blast radius before they touch a line of code — rendered as one standalone HTML page.
A penetration-testing methodology delivered as one MCP server: BLITZ maps the attack surface, EAGLE-EYE traces source-to-sink, STRIKE live-verifies every finding before it is reported.
Anthropic's official Claude Cowork plugin for financial advisors — eight ready-to-run compliance-aware workflows over the firm's own connectors.
Answers macOS Spotlight and Siri prompts with your signed-in Claude Code account — an experimental macOS 27 model-delegation provider.
Anthropic's official "Code with Claude" workshop materials — nine hands-on courses on skills, MCP, evals and Claude Managed Agents.
Anthropic's official K-12 teaching skills — lesson planning, differentiation, prep and formative checks, with an eval framework.
Production-grade engineering skills for AI coding agents — 25 skills and 9 lifecycle slash commands, from spec to ship.
An agent-led search engine scored by upvotes, likes and real money — nineteen sources searched in parallel, synthesized into one brief.
A Claude Code plugin for ADHD brains that remembers what you said but never finished — two-layer capture, session digests, gentle nudges and a focus mode. Yours, not the model's.
Extract your own local chat history from ten AI coding assistants into one normalized JSONL — for fine-tuning, analytics, or backing up years of conversations. Standard library only.
A standalone agent skill for creative 3D visualization — a ten-step workflow, Three.js kits, and a capture helper so the agent inspects the frames it actually rendered.
A local privacy gateway for AI coding assistants — masks credentials, connection strings and PII into structured placeholders on the way out, restores them mid-stream on the way back.
Topic in, narrated explainer video out — a Claude Code / Codex skill that produces full motion-graphics explainers with Remotion, every frame drawn in code.
An open-source orchestrator for running a team of coding agents — Claude Code, Codex, Cursor and more — from your desk, your phone, or any machine.
A long-term memory runtime for AI agents: a markdown store with local ranked retrieval, shared by Claude Code and Codex CLI, no API keys.
An ontology-driven development skill: requirements exploration → seven-model ontology YAML → a runnable management system, with a human sign-off gate at every stage.
An agent skill for building premium scroll-driven websites — design standard, engine, and verification tools in one package.
A macOS notch app that shows how much of each coding assistant's usage limit you've burned — Claude Code, Cursor, Codex, and more at a glance.
Local observability for AI coding agents — a live map of what changed, what's planned, and where your attention is needed.
Shopify's official implementations of Anthropic's commerce-agents blueprint — a storefront shopping agent and a merchant agent against a real store.
Version-by-version teardown of Claude Code's system prompts and 27 built-in tool descriptions, updated with every release.
A single-file CLAUDE.md distilling Karpathy's coding observations into four working rules (211k stars).
The de-facto distribution channel for the open skills ecosystem — `npx skills add`, plus the skills.sh directory.
The highest-signal personal skills collection in the ecosystem (256k stars), organized by failure mode.
Red Hat's "ripgrep for AI context" — a C++23 zero-dependency CLI that also speaks MCP.
One shared session memory for 25+ coding agents — Claude Code, Codex, and Cursor finally remember the same project.
Anthropic's official reference harness for agentic vulnerability discovery — threat modeling, scanning, triage, and patching skills.
A code map of Claude Code's hooks, skills, and MCP surfaces — with official data showing 42% token savings.
Spotify's official Claude Code plugins — Portal catalog workflows plus "shunt", which routes heavy I/O to cheaper models.
A zero-dependency (SQLite) persistent memory layer for agents — 4-tier consolidation with decay-based forgetting (28k stars).
Vercel Labs' Rust-native browser automation CLI — snapshot + ref interaction, with a built-in MCP mode (42k stars).
Anthropic's official blueprint for shopping and merchant agents — prompts, skills, tool contracts, and guardrails defined once.
Anthropic's announcement of Fable 5.1 and Mythos 5.1 — one base model, two guardrail postures.
163 ready-to-run scientific skills (genomics, chemistry, medicine) plus 100+ scientific databases, Agent Skills-compatible.
Exa's official MCP server — live web search, crawling, and deep research wired into Claude.
An open-source, E2E-encrypted mobile and web client for Claude Code and Codex.
Open-source secure sandboxes where AI agents run code — the de facto execution layer for agent apps.
Anthropic's canonical essay on agent design — workflows vs agents, and five composable patterns.
Official Chrome DevTools MCP server — performance tracing, network inspection, and live debugging for coding agents.
The de-facto standard ChatGPT role-prompt collection (f.k.a. Awesome ChatGPT Prompts).
Vercel's handler for spinning up MCP servers on Next.js, Nuxt, Svelte, and more.
HumanLayer's 12 design principles for production-grade LLM agents — the closest thing to a 'best practices' manifesto.
Anthropic's official Claude API quickstarts — agentic customer service, RAG, skills, and more, in working notebooks.
Official curated registry of MCP servers — discovery, versioning, and metadata for the ecosystem.
Turn any GitHub project into an MCP docs hub — no server-side install required.
1000+ tool integrations behind a single SDK — GitHub, Slack, Notion, Gmail, Salesforce, and more.
Official MongoDB-maintained MCP server for Atlas and self-hosted clusters.
Official Sentry MCP server for error monitoring, issue triage, and release health.
Cloudflare's official MCP covering Workers, Observability, Radar, and AI Gateway.
Pythonic framework for building MCP servers — decorators, type hints, and zero boilerplate.
Community-curated collection of Claude prompts.
A composable skill framework that makes Claude Code far more capable.
Anthropic's official TypeScript SDK for building agentic applications.
Python edition of the official Claude Agent SDK.
Anthropic's official agentic coding CLI for the terminal.
Up-to-date documentation MCP server for libraries and SDKs.
The largest curated list of MCP servers — 200+ entries across languages and use cases.
Official GitHub-maintained MCP server for repos, issues, PRs, and code search.
Official docs for Claude mods — plugin event handlers that run inside Claude Code itself, drawing UI and reshaping tool calls.
A Claude Code skill that turns one sentence into a 10-plus-minute faceless investigation video: research, script, TTS, footage, Remotion render, mixing and publishing copy.
Find past Claude Code, Codex and OpenCode conversations from a plain-language description, inside VS Code.
Scan, classify and sync SKILL.md skill libraries across Codex, Claude Code, Copilot, Gemini, Cursor and OpenCode — with an honest A/B/C/D verdict per skill.
A Chinese prose-quality skill that makes AI write modern Chinese worth reading — explicitly not an AI-flavor remover.
An agent plugin that answers one question — should I /compact now? — with a Jev judgment of whether the session sits at a safe boundary.
Three Japanese-language Marp slide skills plus a black-and-cyan theme and five measurement scripts that catch the layout failures eyes miss.
Thirteen free Claude skills that run an Instagram account end to end — nothing posts until you say yes.
A skill that strips the 41 patterns that mark text as AI-written, then runs four gates — ending with a second reviewer — before handing it back.
A local-first job-hunt pipeline that fetches roles straight from ATS APIs, dedupes in SQLite, and lets Claude score each fit 0–100 against your profile.
Exposes locally deployed MCP servers as strict OpenAPI 3.1.0 REST routes, built for the Open WebUI ecosystem.
Mines your own Claude Code and Codex history for the asks you keep retyping and turns each one into a project shortcut like /sent or /project billing.
Self-hosted cognitive memory for coding agents: a Docker stack around a Neo4j knowledge graph that persists decisions and schema facts across Cursor, Claude Code and VS Code.
A free MCP server that hands your agent the Awwwards archive as design inspiration — search by color, tag or tech, see screenshots inline, pull any site's design DNA.
An in-engine Godot plugin that exposes 59 tool families and 1,820+ editor operations over MCP — scene building, tilemaps, physics, shaders and live game control, fully local.
Battle-tested watchdog scripts that keep a Claude Code + Telegram session alive on a VPS — auto-restarts, missed-message recovery and credential refresh, checked every 3 minutes.
Claude Code skills that wait for a third-party PR review and then adjudicate whether its verdict actually means pass — never inferring safety from silence.
A skill that moves the agent's grilling design interview onto a local browser page — question cards with recommendations, per-question threads, one Send button.
A local MCP server that turns TypeSafe's Jev into eight typed judgment tools — verify, screen, find, rerank, classify, decide, compare, extract.
A macOS multicam recorder that hands screen-plus-webcam footage to Claude Code — the agent cuts the video, writes the titles, and publishes it.
A local-first desktop platform where a Kanban board drives role agents — every task is handed to Claude Code on your own machine, column by column.
A self-evolving ontology layer for data agents, exposed through MCP tools and shipped with a Claude Code plugin — from Renmin University's data lab, with an arXiv paper.
A Rust CLI that ranks an agent's installed skills for the current step using TypeSafe's Jev — harness-aware, inspectable, and advisory by design.
A Claude Code plugin marketplace of function-hook mods that draw into the strip above the prompt — a live HUD that spends zero model tokens.
A self-contained pack of 28 engineering-workflow skills that walks an AI coding agent through DEFINE, PLAN, BUILD, VERIFY, REVIEW and SHIP instead of letting it improvise.
An agent skill that turns a topic into a 30-second vertical film where every pixel is drawn in JavaScript and every sound is synthesised — zero media assets.
A local research workspace where Claude Code works beside you: chat, files, PDFs and live Jupyter notebooks in one dock — and the agent can build new panes when none fit.
Compile an agent skill (SKILL.md) into an extended finite state machine, so the order of operations is enforced by the program while models reason inside the states.
Per-turn model routing for Claude Code: TypeSafe's Jev classifies each fresh prompt and a loopback proxy moves you between Haiku, Sonnet and Opus — native picker, sessions and auth untouched.
A self-hosted, versioned skills library for AI agents: immutable revisions, scoped client keys, an HTTP MCP server plus stdio bridge — and it never executes uploaded skill code.
A local MCP server that hands Claude Code (and Codex, Cursor, OpenCode) structured code-quality scores from TypeSafe's Jev — 15 always-on dimensions, no hosted backend, your key stays local.
55 field-tested Claude Code tips across nine sections — sessions, commands, subagents, skills, MCP, hooks, workflow and context — one short recipe each.
Turns vibe coding into high-quality code: an agent skill that writes the rules your repo needs and turns every rule into a gate the agent cannot skip.
A one-file Claude Code skill that heartbeats idle sessions to keep the prompt cache warm — so coming back doesn't re-bill your entire conversation.
/scout turns one topic into a ranked YouTube research workbook — performance, creators, comments, hooks and replay hotspots in one .xlsx.
A nightly-scanned catalog of 50 Claude Code mods — function-hook plugins listed with each one's exact reach: what it hooks, what it can touch.
One desktop app to configure and run the coding-agent CLIs you already have — Claude Code, Codex, Gemini CLI and more — without touching their internals.
A measured, plain-Claude-Code rebuild of the Spotify setup that cut token usage 90%: a cheap model does the reading and the boilerplate, Claude keeps the thinking.
Anthropic's official starter kit for large-scale language migrations — the prompts, templates and scripts behind their documented process.
Anthropic's official gallery of standalone HTML artifacts — code reviews, slide decks, status reports and mini editors with zero build step.
A personal cloud OS for a developer and their AI agents — one Ubuntu server, browser desktop, Claude Code workspaces, 944-app self-hosted store.
MCP server that lets Claude Code drive a live PyMOL session — fetch structures, build selections, adjust representations, measure distances.
MCP server for Shopware 6 — products, orders, customers, stock and documents, plus a one-call health audit and Markdown reports.
Codebase or system description in, a polished interactive system map out — typed JSON IR deterministically compiled to HTML/SVG.
Google's official repo of Agent Skills for Google products — Cloud onboarding recipes, multi-product solution architectures, Agent Platform operations.
Six daily-driver skills from Chinese AI blogger 卡兹克 — goal definition, disk cleanup, AI news digest, doc hygiene, research reports, style-matched long-form writing.
Simon Willison's git-history mirror of Anthropic's published claude.ai system prompts — every revision a dated commit, every change summarized.
Reads your Claude Code and Codex session history and draws what you actually built — every day, task and prompt on one page.
A Claude Code skill that turns a feature description into a clean, abstract product illustration — as real hand-written SVG, not a picture of one.
One shared skill for Codex and Claude Code that turns a product or idea into a complete motion film — ten styles, a local playable gallery, generation via your Higgsfield account.
A hand-checked catalog of SEO agent skills — technical audits, keyword research, content briefs, schema and GEO — with each pack's coverage, supported agents, licence and last-touch date on record.
A hand-checked list of every MCP server that connects an agent to SEO data — Search Console, keywords, backlinks, crawls, SERPs and AI search visibility — with operator, auth needs and last-touch date per entry.
A native macOS workspace that gives Claude Code, Codex, OpenCode and Cursor one shared memory — a local knowledge graph built from your own agent history, searchable everywhere.
A complete open guide plus vault template and agent skills: let an AI agent build and maintain a linked markdown knowledge base for you, in files you own.
A motion-first creative-web Agent Skill — real spring-damper physics, seven runnable cases, automated headless verification against AI slop.
Eleven free MIT-licensed Claude skills that draft, humanize and run a LinkedIn presence — nothing posts until you say yes.
A video-side Claude Code / Codex skill: shot-by-shot film analysis where ffmpeg measures every cut and the report must clear 14 quality gates.
A privacy-first, client-side trace workbench — turn Claude Code, Codex and OTLP logs into a searchable timeline, recorded metrics and side-by-side comparisons. No backend, no telemetry.
punkpeye's curated list of remote MCP servers — hosted endpoints you connect to over a URL, each verified to answer an MCP initialize handshake before it's listed.
One self-contained prompt that turns any capable AI into a refund assistant — a polite policy-based letter first, a statute-backed escalation ladder if they stall.
A skill that generates Chinese official-document DOCX to the GB/T 9704-2012 standard — and machine-verifies the layout.
A /debuzz skill that hands Claude's last answer to Gemini (via the Antigravity CLI) and prints the plain-English translation verbatim.
A ~150-line skill that stops coding agents from servicing their own bookkeeping instead of shipping the product.
Generate and edit GPT images from Claude Code or Codex through your ChatGPT subscription — not the separately billed Images API.
A local A-share quotes/quant/trading plugin for Claude Code, Codex and other agents via the FQGate gateway — unofficial, data from 同花顺.
A source-aware China travel knowledge base and Agent Skill — CLI, HTTP API and MCP server over one JSON data layer, every fact source-linked.
Serve Qwen3.8-27B on Kaggle's free TPU v5e-8 behind an OpenAI-compatible endpoint you can plug into Claude Code, Codex CLI, or opencode.
A read-only one-command inventory of every AI agent, harness, MCP server, plugin, and AI extension on a machine, with plain-language exposure labels.
A local, read-only MCP server that lets Claude Desktop read your Claude Code session history from ~/.claude, with secret redaction.
A curated, link-checked catalog of 16 MCP servers for French data, laws, and services — open data, legislation, government services, and transport.
An agent skill for browser motion graphics — openers, promos, idents, kinetic typography, and explainers that render as a single autoplay index.html.
An agent skill that takes a hardware idea to a board you can order — schematic, sourcing, layout, routing, verification — driving EasyEDA Pro over MCP, with the purchase staged to the pre-payment page.
An agent skill that turns Claude Code into a voiceover-video studio: word-level sync, 108 motion recipes, three-stage acceptance.
Open-source SEO + GEO skills for Claude, running on your real Search Console, GA4, and ads data through a free MCP connector.
Git-native persistent memory for AI agents — plain Markdown knowledge bundles conforming to Google's OKF v0.2, with a Go CLI and stdio MCP server.
A Datalog engine for LLM agent memory — memory as a deductive database with provenance, not a bigger pile of embeddings.
A Claude Code skill that measures cyclomatic complexity and refactors the worst hotspots — without gaming the metric.
A self-hosted, read-only MCP server for your own bank accounts — 2,700+ European banks over PSD2, no payments, one user.
A local password manager and stdio MCP relay that lets agents use credentials without exposing hidden plaintext.
A curated list of exemplary CLAUDE.md files from real open-source projects.
A continuously updated archive of extracted system prompts, fastest to cover the newest Claude models (Fable 5.1, Opus 5, Claude Design).
A complete, working CLAUDE.md memory-bank starter: config plus the hooks, commands, and subagents that make it run.
A plugin marketplace with its own curated collections, plus a discovery hub for the wider Claude Code extension ecosystem.
Garry Tan's 23-specialist Claude Code factory — CEO, Designer, QA and more — wired into a sprint from idea to shipped PR.
Disk-persistent task plans (task_plan / findings / progress) re-injected every turn — survives /clear and compaction.
Tencent's open-source RTL waveform-debugging MCP — 34 tools that let Claude read and drive chip simulations.
A Rust single-binary web fetch/search/crawl MCP — no API keys, no accounts.
A meta-harness over Claude Code, Codex, Cursor, OpenCode, Hermes, Pi, and more — YAML-defined agents with cross-vendor subagent review.
Anthropic's official skill for taking an agent from interview to launch on Claude Managed Agents.
Trail of Bits' one-shot microVM runner — Claude Code, Codex, or Grok Build inside Firecracker/Lima, thrown away after each session.
The official "Skills over MCP" incubation repo — distributing skills through MCP Resources.
Turns codebases, docs, PDFs, and SQL into a queryable knowledge graph — no vector DB required (115k stars; read the caveats).
Reads your local agent session logs and emits a "work receipt" — what was done, what it cost, how it was verified.
Anthropic's on-device C2PA content-credential detector — check whether a file was issued by Claude, without uploading it.
An interactive cluster analysis of 467K pull requests — tracking when Claude's load-bearing words entered GitHub and spread.
A native x64dbg plugin that exposes the debugger's full toolset to any MCP client — reverse engineering with an agent in the loop.
"The AI proposes. The bytes decide." — a binary-analysis anti-hallucination verification layer, callable as an MCP server.
Guizang's art-direction skill: turn architecture, street, and food photos into editorial-grade posters.
A Claude Code skill for boardroom-quality decks: a ~110-rule slide rulebook, a machine rule-checker, and a 62-part HTML slide library.
Anthropic's Fable 5.1 prompting prescriptions packaged as a zero-config hook plugin (KO/EN/ZH).
Anthropic Academy's official course on rebuilding the SDLC around AI — plan, design, build, test, deploy, maintain.
16,893 sessions of experiment data on how Claude Code, Codex, and Cursor actually choose third-party tools.
An open-source multi-provider AI gateway for Claude Code — model routing, fallbacks, and local models behind one endpoint.
A Claude Code skill that applies Refactoring UI's concrete design rules — hierarchy by weight, depth by emulated light.
A Claude Code skill that audits and optimizes five visibility lanes: SEO, AEO, GEO, LLMO, and Naver's NEO.
An open-source AI-coworker platform: coding agents with their own cloud computer, delivering finished work.
Neon's official MCP server — create databases, branch data, and run SQL from Claude.
An agent company for Claude Code — 16 departments and 172 skills, installed as plugins.
Watch a Claude Code session as a live flow graph — in your terminal or browser.
A paste-in RULES.md contract that stops coding agents from over-defending their changes.
Grafana's official MCP server — query metrics, search dashboards, and manage incidents from Claude.
Tavily's official MCP server — production real-time search, extract, map, and crawl for agents.
ClickHouse's official MCP server — let Claude run SQL against your OLAP data.
Search and generate production-ready React + Tailwind UI components from inside Claude Code.
57 production-ready slash commands (15 workflows, 42 tools) for Claude Code, by wshobson.
A portable Agent Skill that strips AI-sounding patterns from writing — fiction and professional docs alike.
A curated directory of Claude Code plugins, skills, and MCP servers, ranked by real usage.
Kanban orchestration for 10+ AI coding agents — plan, run, and review their work on one board.
Team chat where AI agents are first-class participants — personas, memory, Kanban, and real email.
Observability for AI agents — session replay, LLM cost tracking, and monitoring in a few lines.
An open-source testing and evaluation platform for MCP servers — Postman for the MCP ecosystem.
Anthropic's official documentation for Agent Skills — SKILL.md structure, loading, and progressive disclosure.
Anthropic's follow-up essay on context engineering — curation, compression, and attention budgets.
The official guide to wiring MCP servers into Claude Code — scopes, transports, and auth.
Framelink's Figma MCP server — feed real design layout data to Claude for design-to-code.
Microsoft's official MCP server catalog — Azure, Learn, Fabric, Sentinel, Teams, and more.
Anthropic's official plugin directory for Claude Code — curated plugins installable via /plugin.
Anthropic's official plugin collection for knowledge work — productivity, sales, finance, legal, and data workflows.
GitHub's official spec-driven development toolkit — specify, plan, tasks, implement with any AI agent.
Lightweight spec-driven development for AI coding agents — the minimal alternative to Spec Kit.
Agile AI-driven development with agent roles — analyst, PM, architect, dev, and QA as a virtual team.
PRD → tasks → subtasks task management for AI agents, with CLI and MCP server.
Project management for Claude Code — PRDs become GitHub Issues, executed by parallel agents on worktrees.
A spec-driven system that gives AI coding agents your standards, context, and workflow.
TypeScript-first agent framework — agents, workflows, memory, evals, and MCP in one package.
Hugging Face's minimal agent library — agents that think in Python code.
Type-safe agent framework from the Pydantic team — structured outputs and typed tools for Claude.
Web and mobile interface for managing Claude Code sessions remotely.
Dagger's containerized development environments for coding agents — every agent gets its own container and branch.
Anthropic's official demo apps for the Claude Agent SDK — email, research, and chat agents.
Token usage and cost analytics for Claude Code, straight from your local logs.
The leading third-party MCP server registry — thousands of servers with managed install and OAuth.
Anthropic's official 9-chapter interactive prompt engineering course, runnable in your browser.
Anthropic's engineering blog — how Claude Code, agents, and harnesses are actually built.
Anthropic's official GitHub Action that reviews PRs for security vulnerabilities with Claude.
AI gateway with a Rust core — call Claude and 100+ providers through one OpenAI-format API.
The open source coding agent — terminal-first, model-agnostic, with shareable session links.
Persistent memory for Claude Code — captures sessions, compresses them, and reinjects context later.
Hugging Face's free course on building AI agents — from fundamentals to a certified final project.
Microsoft's 21-lesson course on building with generative AI — from prompting to agents.
Spec-driven development as an MCP server — requirements → design → tasks, with a real-time dashboard.
Official reference MCP server for working with local Git repositories.
Official reference MCP server for current time and timezone conversion.
Linear's hosted MCP server for issues, projects, and comments — OAuth, no local install.
Elastic's official MCP server for natural-language Elasticsearch querying (deprecated — see Agent Builder).
Docker's official CLI plugin that runs MCP servers as isolated containers behind one gateway.
MCP server that lets Claude drive the iOS Simulator — UI inspection, taps, screenshots.
Multi-model orchestration MCP (formerly Zen MCP) — let Claude call Gemini, GPT, and local models for second opinions and consensus.
93 plugins, 202 agents, and 181 skills for Claude Code — also consumed natively by Codex CLI, Cursor, and Copilot.
A configuration framework that gives Claude Code 30+ slash commands, cognitive personas, and development methodologies.
A curated collection of 100+ specialized Claude Code subagents across languages, DevOps, and QA.
The original agent meta-harness for Claude Code — self-organizing swarms with adaptive memory.
Manage multiple AI terminal agents (Claude Code, Codex, Gemini, Aider) in parallel tmux sessions.
Open-source, extensible AI agent that installs, executes, edits, and tests with any LLM.
Role-playing multi-agent orchestration framework — crews of agents collaborating on tasks.
Anthropic's official prompting guidance — clear instructions, examples, XML structure, and thinking.
The official Java SDK for MCP, maintained in collaboration with the Spring AI team.
The TypeScript toolkit for AI apps and agents — with first-class Anthropic/Claude support.
The SDK for browser agents — Playwright-style API with act, extract, and observe primitives.
Hugging Face's free, open course on building MCP servers, clients, and agents.
Anthropic's official courses and certifications for the Claude API, Agent skills, and Claude Code.
Curated prompts, frameworks, and papers with an engineering bias — 20 prompt domains plus eval and red-team sections.
Organized library of system prompts, custom instructions, jailbreak and protection prompts across 25+ vendors.
Reverse-engineered system prompts from production AI tools — for prompt engineering research.
A curated collection of Claude Code slash commands and custom slash-command templates.
Run Claude Code remotely via a Telegram bot — pipe messages from phone to your dev machine.
disler's production-grade Claude Code hooks — real configurations with explanations.
Stripe's official AI toolkit — hosted MCP server at mcp.stripe.com plus billing SDKs for LLM apps.
ElevenLabs' official MCP server for text-to-speech, voice cloning, and audio transcription.
488+ standardized AI prompts plus 260+ Claude Code skills and 7 automation hooks.
Leaked system prompts from widely used AI products — for prompt engineering research.
Official Swift SDK for building MCP servers and clients on Apple platforms.
Official Kotlin SDK for MCP — built with JetBrains, targeting JVM/Android/server-side Kotlin.
Official C# SDK for MCP, hosted by Microsoft and targeting .NET 8+.
A structured, modular prompt-template language designed for LLM agents — Chinese-origin community standard.
Microsoft's free 18-lesson course on building AI agents — covers GenAI design patterns from basics.
Tencent's open-source agent framework — single-agent, multi-agent, and deep-research modes.
Linux Foundation-hosted framework for building production-ready multi-agent systems in Python or TypeScript.
Cloudflare's open-source SDK for building stateful AI agents on Workers + Durable Objects.
Browser-Use lets LLMs drive a real browser — navigate, click, fill forms, extract data.
Bash / JS / Python functions exposed as LLM tools, with two-way MCP support — a runtime layer for LLM agents.
A terminal utility that turns plain-language descriptions into shell commands via a CTRL+F widget.
Official Containers project MCP server for kubectl-grade operations against any k8s cluster.
Stacklok's MCP runtime — secure proxy, registry client, and deployment layer for MCP servers.
Public collection of leaked system prompts from major AI products — research-only.
Anthropic's experimental research on how agents behave under prompt injection and conflicting incentives.
MCP server for Obsidian vaults — search, read, and write notes via STDIO or HTTP.
Curated bundle of Claude Code skills covering code review, TDD, debugging, and brainstorming workflows.
A 2,000+-skill catalog that grew a control plane: AAS Core lets your agent search the local catalog read-only and validates the exact stack it composes.
Graph-based orchestration framework for stateful, multi-agent LLM workflows.
MCP-first agent framework with Temporal-backed durable execution and composable workflow primitives.
Multi-agent framework that simulates a software company — ProductManager, Architect, Engineer roles.
Official Rust SDK for building MCP servers and clients, published by modelcontextprotocol.
Official Go SDK for MCP — built with Google as a reference implementation partner.
Official Anthropic PHP SDK for the Claude API.
Microsoft's library for replacing prompt strings with type-safe TypeScript schemas.
Official Redis-maintained MCP server for strings, JSON, hashes, and vector search.
Serverless-friendly Redis MCP via Upstash HTTP, ideal for edge deployments.
Modern Postgres MCP with schema introspection, index tuning, and EXPLAIN plan analysis.
AWS Labs umbrella repository for MCP servers covering Lambda, CDK, Pricing, and Diagnostics.
Community-maintained reference MCP for Jira issues, Confluence pages, and Atlassian search.
MCP server for Airtable bases, tables, records, and views.
Cloud browser automation MCP powered by Browserbase and Stagehand.
Community Go SDK for MCP servers and clients.
Official Anthropic Go SDK for the Claude API.
Official Anthropic Java SDK for the Claude API.
Official Anthropic Ruby SDK for the Claude API.
Official Anthropic C# / .NET SDK for the Claude API.
Official walk-through for writing an MCP server in TypeScript.
Curated list of Claude tools, SDKs, prompts, and tutorials.
Anthropic's guide to getting the most out of Claude Code.
Official Claude docs covering tools, prompts, and best practices.
Official MCP documentation, quickstarts, and examples.
The official specification of the Model Context Protocol.
Anthropic's official prompt engineering documentation.
Comprehensive prompt engineering guide by DAIR.AI.
Official courses on prompting, tool use, and real-world Claude apps.
Official Anthropic notebooks for Claude — prompting, tool use, retrieval, and agent patterns.
Built-in assistant and agent features in the Zed editor.
Open-source terminal AI coding agent with a TUI.
Original open-source Claude coding plugin for VS Code.
Self-hosted developer control center for coding agents and automations (beta) — the repo that began as OpenDevin.
Terminal-based AI pair programmer that edits git-tracked files.
Fork of Cline with custom modes and rich MCP support.
Autonomous AI coding agent inside VS Code.
Open-source AI coding assistant plugin for VS Code and JetBrains.
Run specialized Claude Code sub-agents in parallel from one session.
Tools, skills, plugins, and MCP servers for agentic coding.
Anthropic's official GitHub Action that runs Claude Code in CI.
Route Claude Code requests to any LLM provider you choose.
Reusable .claude templates, commands, and hooks for common stacks.
Hand-picked collection of Claude Code resources — skills, agents, status lines, tooling, and plugins.
Official public repository of example Agent Skills for Claude.
Official TypeScript SDK for building MCP servers and clients.
Official Python SDK for building MCP servers and clients.
Official Python client for the Anthropic API.
Official TypeScript client for the Anthropic API.
Official Notion MCP server for pages, databases, and blocks.
Official Supabase MCP server for Postgres, Auth, and Storage.
Turn any website into LLM-ready markdown via Firecrawl.
Microsoft's Playwright MCP server for browser automation.
Official interactive debugger for MCP servers — list tools, call them, and inspect JSON-RPC frames.
Bridge between Streamable HTTP and stdio MCP transports.
Reference MCP server for GitLab projects and merge requests.
Reference MCP server for searching and reading Google Drive.
Reference MCP server for Google Maps directions and places.
Official reference server for step-by-step dynamic reasoning.
Official Brave Search MCP server for web and local search.
Official reference MCP server for fetching web content.
Official reference MCP server for a persistent knowledge graph.
Community Slack MCP server for channels, messages, and reactions.
Reference MCP server for browser automation via Puppeteer.
Reference MCP server with read-only access to PostgreSQL.
Reference MCP server for SQLite databases.
Official reference MCP server for read/write access to the local filesystem.