MCP · A2A · A2UI · AG-UI · AP2 · x402 · Skills

The protocols that turned AI models into an agent economy

In 20 months the agentic stack went from zero to standardized: one protocol for tools, one for agent-to-agent collaboration, one for interfaces, two for payments — and an open format for packaging expertise. This is the deep, visual guide to the state of the art, with live adoption data — plus a hands-on case study of one real working setup.

7
core protocols, deeply explained
383,206 ★
OpenClaw — most-starred agent repo
161,844 ★
Agent Skills open format (Anthropic)
88,558 ★
MCP servers repo — live GitHub data
The big picture

One stack, six layers

Every acronym in agentic AI fits on this map. The model reasons at the bottom; each layer above gives it a new power — knowledge, hands, colleagues, a face, and finally a wallet. Every layer now has an open protocol.

MONEYAP2 (mandates, fiat rails) · x402 (HTTP-native stablecoin micropayments)agents that can payHUMANS & UIA2UI (declarative agent-rendered UI) · AG-UI (event stream to your frontend)agents you can seeAGENT ↔ AGENTA2A (Agent Cards, tasks, artifacts) · ANP · AGNTCYagents that collaborateAGENT ↔ WORLDMCP (tools, resources, prompts) · function calling · REST/GraphQL APIsagents that actEXPERTISEAgent Skills (SKILL.md) · AGENTS.md · system prompts · memoryagents that know howMODELLLM reasoning loop — plan → act → observe → reflect (ReAct)the engine
For new learners — the 5-minute mental model

Tools, APIs, MCP, Skills — what's actually the difference?

The framing top explainers converge on: tools are the hands, MCP is the port, skills are the playbooks, protocols are the language. An agent is a model given agency (it decides), autonomy (it acts without step-by-step supervision), and self-governance (it stays within your rules) — the three properties Biswas & Talukdar use to define agentic systems.

🔧

Tools the hands

A tool is one function the model may call — search_flights(date, city), run_sql(query). The model reads the tool’s schema, decides when to call it, and gets structured results back. Tool calling is the atom of agency: without tools, an LLM can only talk.

🌐

APIs the raw wiring

APIs are how software talks to software — REST, GraphQL, gRPC. Agents consume them through tools. The problem: every API is different, so every integration was custom glue code. That N×M explosion is exactly what MCP was invented to kill.

🔌

MCP the universal port

The Model Context Protocol standardizes how models discover and call tools, read resources, and reuse prompts. Build one MCP server for your database and every MCP client — Claude, ChatGPT, Cursor, your own app — can use it. USB-C for AI.

📘

Skills the playbooks

A skill is a folder with a SKILL.md — instructions, scripts, references — that teaches an agent how to do something well: build slide decks, review contracts, deploy to Vercel. Loaded only when relevant (progressive disclosure), so 500 skills cost almost nothing until one is needed.

🤖

Subagents the team

Agents that spawn specialized agents — a planner, a coder, a reviewer — each with its own context and tools. The book calls this the Coordinator–Worker–Delegator pattern: one orchestrator, many focused workers, results merged back.

🔁

The agent loop the heartbeat

Every agent runs the same cycle: reason about the goal, act with a tool, observe the result, reflect, repeat (ReAct). Reflection and introspection — the agent critiquing its own output — is what separates an agent from an autocomplete.

Cheat sheet: which piece do you need?

One decision table instead of forty tutorials.

You want the model to…Reach for…Example
Do one concrete actiona Toolsend_email(to, subject, body)
Connect to a whole system, reusablyan MCP serverSupabase MCP: 20+ tools from one connection
Follow your procedure & standardsa Skillslides skill: how YOU build decks
Handle a big task in parallelSubagentsplanner + coder + reviewer
Talk to someone else’s agentA2Ayour agent hires a travel agent’s agent
Show real UI, not text wallsA2UI / AG-UIdate pickers, live progress, approvals
Spend money safelyAP2 / x402signed mandates; per-call micropayments
Deep dives

The seven protocols, one at a time

Each layer of the stack got its own open standard between late 2024 and late 2025. Pick a protocol — every panel has the plain-English version, the architecture, live adoption numbers, and links to the primary sources.

MCPModel Context Protocol

Agent ↔ Tools & Data

Anthropic → open ecosystem (OpenAI, Google, Microsoft adopted) · Born Nov 25, 2024

The open standard that connects an AI model to the outside world: files, databases, SaaS apps, browsers — anything. One protocol instead of an N×M mess of custom integrations.

Explain it like I’m new hereMCP is the USB-C port of AI. Before USB-C, every device needed its own cable; before MCP, every AI app needed a custom integration for every tool. Now you build one MCP server and every MCP-capable model can plug into it.
Host appClaude · ChatGPT · IDEMCP clientMCP serverGitHubMCP serverPostgresMCP serverBrowserJSON-RPC 2.0tools · resources · promptsReal worldAPIs · data · files

How it works

  • Client–server over JSON-RPC 2.0: your AI app (host) runs MCP clients that connect to MCP servers exposing capabilities.
  • Three primitives — Tools (actions the model can call), Resources (data it can read), Prompts (reusable templates).
  • Two transports: stdio for local servers, Streamable HTTP for remote ones, with OAuth 2.1 for authorization.
  • The model discovers what a server offers at runtime — no hardcoding. New server, new powers, zero code changes.

By the numbers

88,558 ★
servers repo stars (live)
10+
official SDKs (Python, TS, C#, Java, Kotlin…)
Thousands
of community MCP servers + official registry
Mar 2025
OpenAI adopted; Google & Microsoft followed

Moving parts

  • Host app (Claude, ChatGPT, IDEs)
  • MCP client (one per connection)
  • MCP server (wraps a tool/data source)
  • JSON-RPC 2.0 messages over stdio / HTTP

The wider landscape

Standards worth knowing that orbit the core seven.

Importance · relevance · popularity

Adoption, in real numbers

Popularity is measurable: repository gravity, partner counts, governance moves. Here is where each protocol actually stands — plus what the analysts project.

$52.6B
projected AI agents market by 2030 (from $7.8B in 2025) — CAGR 46.3%
MarketsandMarkets
33%
of enterprise software applications will include agentic AI by 2028 (from <1% in 2024)
Gartner
15%
of day-to-day work decisions made autonomously via agentic AI by 2028 (from 0% in 2024)
Gartner
79%
of organizations have adopted AI agents or plan to (35% already adopted, +44% planning)
MIT Sloan Mgmt Review / BCG

Ecosystem gravity — GitHub stars per protocol / SDK repo

Fetched live from the GitHub API on July 17, 2026. Linear scale — the disparity is the story.

Agent Skills (anthropics/skills)
161,844
MCP servers
88,558
OpenAI Agents SDK
27,958
A2A spec
24,819
A2UI (Google)
15,778
AG-UI
14,780
MCP spec
8,614
Claude Agent SDK (Py)
7,647
AP2
3,116
ANP
1,352
ACP (dormant)
1,017

Read: packaged expertise (skills) and tool connectivity (MCP) dominate developer attention; agent-to-agent and payments layers are earlier on the curve. ACP is dormant — it merged into A2A.

20 months from zero to standard

Every layer of the stack got an open protocol in under two years.

Nov 2024Anthropic open-sources MCPMar 2025OpenAI adopts MCP — de factostandard momentApr 2025Google launches A2A with 50+partnersMay 2025AG-UI (CopilotKit) + x402(Coinbase) launchJun 2025A2A donated to the LinuxFoundationSep 2025AP2 launches with 60+ paymentpartners; MCP Registry opensOct 2025Anthropic ships Claude Skills(Agent Skills format)Dec 2025Google open-sources A2UI;skills ecosystem explodes2026Consolidation: MCP + A2A +skills emerge as the core stack
State of the art — the ecosystem, ranked

🏆 The Top 20 AI Agent Arsenal (2026)

The most-starred tools, skill packs and harnesses in the global agent ecosystem — every repository verified against the GitHub API on July 17, 2026, and ranked by real star counts, not by the stale numbers circulating in listicles.

#Tool / Skill packStars (live)Primary utilityRepository
1OpenClaw383,206The #1 open-source personal AI assistant — any OS, any platform, the lobster way 🦞.openclaw/openclaw
2Superpowers256,206Agentic skills framework enforcing TDD, brainstorming-first and disciplined planning.obra/superpowers
3Everything Claude Code (ECC)230,454Agent harness optimization: skills, instincts, memory, security, hooks. (Repo renamed from everything-claude-code.)affaan-m/ECC
4n8n196,757Fair-code workflow automation with native AI agent nodes — the no-code gateway to agentic systems.n8n-io/n8n
5Ollama176,293The backbone for running frontier open models locally (Kimi, GLM, DeepSeek, Qwen…).ollama/ollama
6Firecrawl152,147The API to search, scrape and turn any website into LLM-clean Markdown at scale.firecrawl/firecrawl
7Dify149,118Production-ready platform for building and managing agentic workflow backends.langgenius/dify
8Open WebUI145,713The most popular local-first chat interface for LLMs (Ollama, OpenAI-compatible).open-webui/open-webui
9UI UX Pro Max106,748Design-intelligence skill for building UIs that don’t look AI-generated.nextlevelbuilder/ui-ux-pro-max-skill
10Scrapling69,819Adaptive web scraping framework that auto-adjusts when sites change.D4Vinci/Scrapling
11Get Shit Done (GSD)64,761Meta-prompting, context-engineering and spec-driven development framework.gsd-build/get-shit-done
12Awesome Claude Code50,210The master directory of skills, hooks and resources for Claude Code.hesreallyhim/awesome-claude-code
13Impeccable47,431The design language that makes your agent better at design (/audit, /polish).pbakaus/impeccable
14Claude Skills (Fullstack)10,61566 specialized full-stack skills — framework-specific deep knowledge packs.Jeffallan/claude-skills
15Magic MCP5,452v0-style UI component generation injected straight into your IDE via MCP.21st-dev/magic-mcp
16Obsidian Wiki2,891Framework for agents to build and maintain a digital brain in Obsidian.Ar9av/obsidian-wiki
17TÂCHES CC Resources1,973Curated custom Claude Code resources; meta-skill auditing against prompt bloat.glittercowboy/taches-cc-resources
18Claude Pace214Statusline rate-limit tracker with pace-aware quota monitoring. Pure Bash.Astro-Han/claude-pace
19Stitch Kit35Design-to-code bridge on Google’s Stitch engine — 35 skills, 7 framework targets.gabelul/stitch-kit
20notebooklm-py4Full Python API access to NotebookLM beyond the official UI.patriarchal-boothose896/notebooklm-py

Corrections vs. lists circulating online: everything-claude-code is now affaan-m/ECC, Firecrawl moved to the firecrawl org, Magic MCP lives at 21st-dev/magic-mcp, and most star counts out there are months stale — OpenClaw alone gained six figures. Always verify against the API.

Hands-on — distilled from the best tutorials

Use them today: skills & MCP in three moves each

The two pieces you can adopt this afternoon. No framework, no PhD — a folder and a config entry.

📘 Add a skill to Claude Code

Create the folder

A skill is just a directory with a SKILL.md inside.

mkdir -p ~/.claude/skills/deck-builder

Write SKILL.md — frontmatter + instructions

The description decides when the agent loads it. That one line is 80% of the craft.

---
name: deck-builder
description: Use when building slide decks.
  Applies my structure, brand and chart rules.
---
# Deck Builder
1. Ask for audience + duration first.
2. One idea per slide, max 12 slides...

Just ask

Say "build me a deck about X" — the agent sees the skill's description, loads the full instructions only then. That's progressive disclosure: hundreds of installed skills cost ~40 tokens each until one activates.

🔌 Connect an MCP server

Pick a server

Thousands exist: GitHub, Supabase, Playwright, Slack, Stripe, Figma… Browse the official registry or the servers repo.

# one command in Claude Code:
claude mcp add github \
  -- npx -y @modelcontextprotocol/server-github

Authorize it

Remote servers use OAuth 2.1 — you approve scopes once; local servers run on stdio with your credentials in env vars, never in code.

The model discovers the rest

Tools, resources and prompts are self-describing. Ask "list my open PRs" and the model finds search_pull_requests on its own. New server version, new tools — zero changes on your side.

The rule of thumb the tutorials agree on

MCP for connections, skills for conventions. MCP gives the agent access to systems (what it can touch); skills encode how your team wants things done (how it should behave). They compose: a deploy skill that calls the Vercel MCP server is the whole stack in one workflow.

Hands-on lab — create your first skill

Build a real SKILL.md, right here

A skill is a folder with a SKILL.md file inside. Six parts do all the work — learn what each one is for, then fill in the form and copy a working skill you can install in one minute.

🏷️

1 · Frontmatter: name required

Lowercase letters, numbers and hyphens only (max 64 chars). It becomes the folder name and how the agent refers to the skill — deck-builder, not "My Deck Builder!".

🎯

2 · Frontmatter: description the trigger

The single most important line (max ~1024 chars). It is the ONLY part the agent always sees, and it decides when the skill activates. Formula: what it does + when to use it — "Use when the user asks for a slide deck. Applies my structure, brand and chart rules."

📋

3 · Body: When to Use scope

A short section listing trigger cases and — just as important — when NOT to use the skill. This stops the agent from over-applying it to everything.

⚙️

4 · Body: Instructions the playbook

Numbered steps written like you would brief a new teammate: exact order, decisions to make, quality bar, what "done" means. Concrete beats clever — "max 12 slides, one idea each" outperforms "make it good".

💡

5 · Body: Examples & Guardrails calibration

One or two input → expected-output examples calibrate the agent better than paragraphs of prose. Guardrails list hard limits: never invent data, always ask audience first, keep files under 800 lines.

📁

6 · Optional: references/ & scripts/ on demand

Big reference docs go in references/*.md and executable helpers in scripts/ — the agent reads them only when needed. This is level 3 of progressive disclosure: your SKILL.md stays lean.

How the agent actually reads it — progressive disclosure

Level 1: at session start the agent sees only name + description (~40 tokens per skill — that's why 500 installed skills cost almost nothing). Level 2: when your request matches the description, the full SKILL.md body loads. Level 3: references/ and scripts/ are read only if the task needs them. Write each level accordingly: razor-sharp description, lean body, deep references.

live preview · ~211 tokens~/.claude/skills/deck-builder/SKILL.md
---
name: deck-builder
description: Builds slide decks with my structure, brand colors and chart rules. Use when the user asks for a presentation, slide deck, or pitch.
---

# Deck Builder

## When to use

- Use when the user asks for a presentation, slide deck, or pitch.
- Do NOT use for unrelated requests — stay in scope.

## Instructions

1. Ask for the audience, duration and goal before writing anything
2. Outline max 12 slides — one idea per slide
3. Write the deck as a single HTML file with speaker notes
4. End with a summary slide and one clear call to action

## Example

**User:** "Build me a 10-minute deck about AI agents for HR directors"
**Agent:** applies every step above, in order, before delivering.

## Guardrails

- Never invent statistics — ask for sources or mark as placeholder
- Keep text per slide under 40 words

Install it (Claude Code)

mkdir -p ~/.claude/skills/deck-builder
# paste the copied content into:
#   ~/.claude/skills/deck-builder/SKILL.md
# then just ask: "Build me a 10-minute deck about AI agents for HR…"
Primary sources

The library: papers, specs & guides

Everything on this site traces back to a primary source. Start with the two surveys, then go protocol by protocol.

📄 Papers & books

📚 Specs & engineering guides

Sample of use — a case study, not a benchmark

Case study: one practitioner's skill library, measured

Everything above is the global state of the art. This section is different: a concrete example of what one real working setup looks like — a live scan of the skills installed on the author's machine (1,654 SKILL.md files, deduplicated to 523 unique skills, including 13 of the Top 20 arsenal). Use it to calibrate what "adopting the stack" means in practice, not as an industry reference.

523
unique skills installed
1,654
SKILL.md files on disk (incl. duplicates)
~1.18M
estimated tokens of packaged expertise
~2,253
avg tokens per skill

Top 10 heaviest skills by token weight

Estimated as SKILL.md bytes ÷ 4 · scanned 2026-07-17

graphify
14,375
ui-ux-pro-max
11,194
skill-creator
8,292
workflow
8,095
quality-nonconformance
7,597
energy-procurement
7,519
feature-usage-feed
7,405
customs-trade-compliance
7,261
production-scheduling
7,143
notebooklm-py
6,475

Most used in practice

Curated from real workflows: teaching decks, research, shipping Vercel apps.

slidesHTML slide decks for teaching and client workshopsteaching
deep-researchMulti-source cited research via Firecrawl + Exaresearch
market-researchMarket sizing & competitive analysis for AplicAppresearch
frontend-designDistinctive UI for every app shippedbuilding
graphifyTurns any input into knowledge graphs (his largest skill: ~14.4k tokens)building
copywritingLanding copy that convertsmarketing
prompt-engineerSharper prompts for agent pipelinesagents
create-agent-skillsMeta: skills that write skillsagents
vercel:deployShip to production from the terminalbuilding
token-budget-advisorCustom-built: controls response depth & costagents