mcp-auth-oauth

mcp-auth-oauth is a skill for Claude Code from latestaiagents/agent-skills. It costs 108 tokens per session (1,647 once invoked), scanned A, original, MIT.

A guide to protecting remote MCP servers with OAuth 2.1 and PKCE, which let users sign in and grant limited access without sharing permanent credentials.

In plain words
What is it for?
Use it to add authentication to a multi-user MCP server, migrate from static tokens, or investigate authorization failures.
Why use it?
It replaces hard-coded bearer tokens with a user-based login flow, token renewal, and permission scopes.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions Claude Code.

Part of the mcp-mastery plugin — 7 skills shipped together , and of latestaiagents

Good fit Use it to add authentication to a multi-user MCP server, migrate from static tokens, or investigate authorization failures.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/latestaiagents/agent-skills/mcp-auth-oauth
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add latestaiagents/agent-skills --skill mcp-auth-oauth
Clone the repo
git clone --depth 1 https://github.com/latestaiagents/agent-skills

Made for: Claude Code.

Or install mcp-mastery, the plugin that ships this one along with the rest of its 7 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp-auth-oauth

README.md
[![agentmods](https://agentmods.dev/badge/skills/latestaiagents/agent-skills/mcp-auth-oauth/github.svg)](https://agentmods.dev/skills/latestaiagents/agent-skills/mcp-auth-oauth)
Your own site
<a href="https://agentmods.dev/skills/latestaiagents/agent-skills/mcp-auth-oauth"><img src="https://agentmods.dev/badge/skills/latestaiagents/agent-skills/mcp-auth-oauth/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for mcp-auth-oauth

Your own site · 80×15
<a href="https://agentmods.dev/skills/latestaiagents/agent-skills/mcp-auth-oauth"><img src="https://agentmods.dev/badge/skills/latestaiagents/agent-skills/mcp-auth-oauth.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 108 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,647 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00108 $0.01647
Opus 5 $0.00054 $0.00823
Sonnet 5 $0.00022 $0.00329
Haiku 4.5 $0.00011 $0.00165

Measured 9d ago against content hash 6636336c7daf, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

mcp-auth-oauth scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/mcp-mastery/mcp-auth-oauth/SKILL.md · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP Auth — OAuth 2.1

Remote MCP servers use OAuth 2.1 + PKCE with Dynamic Client Registration. Any spec-compliant MCP client — Claude Desktop, Claude Code, Cursor, ChatGPT Apps — can authenticate users without hardcoded client IDs.

When to Use

  • Building a remote MCP server that serves multiple users
  • Migrating a static-bearer-token server to per-user auth
  • Debugging "401 Unauthorized" or "failed to authenticate" in an MCP client
  • Adding scope-based permission to an MCP server

The Flow (RFC 8414 + RFC 7591 + PKCE)

  1. Client connects to MCP server → server returns 401 with WWW-Authenticate pointing to the authorization server metadata URL
  2. Client fetches /.well-known/oauth-authorization-server → gets authorization_endpoint, token_endpoint, registration_endpoint
  3. Client POSTs to registration_endpoint (RFC 7591) → receives a client_id dynamically
  4. Client opens browser to authorization_endpoint with PKCE challenge → user approves
  5. Client exchanges code at token_endpoint → receives access + refresh tokens
  6. Client retries MCP request with Authorization: Bearer <token>

No pre-shared client ID. No manual app registration. The user just clicks "Allow" in a browser.

Server-Side Requirements

1. Return discovery metadata

// GET /.well-known/oauth-authorization-server
app.get("/.well-known/oauth-authorization-server", (req, res) => {
  res.json({
    issuer: "https://mcp.example.com",
    authorization_endpoint: "https://mcp.example.com/oauth/authorize",
    token_endpoint: "https://mcp.example.com/oauth/token",
    registration_endpoint: "https://mcp.example.com/oauth/register",
    scopes_supported: ["read:issues", "write:issues"],
    response_types_supported: ["code"],
    grant_types_supported: ["authorization_code", "refresh_token"],
    code_challenge_methods_supported: ["S256"],
    token_endpoint_auth_methods_supported: ["none"], // public clients
  });
});

2. 401 with WWW-Authenticate

app.use("/mcp", (req, res, next) => {
  const token = req.headers.authorization?.replace("Bearer ", "");
  if (!token || !verifyJwt(token)) {
    res.setHeader(
      "WWW-Authenticate",
      `Bearer resource_metadata="https://mcp.example.com/.well-known/oauth-protected-resource"`,
    );
    return res.status(401).json({ error: "unauthorized" });
  }
  req.user = decodeJwt(token);
  next();
});

Read the full file on GitHub · 174 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 174 lines · 108 tokens per session scan A 6636336c7daf

Subscribe to this mod's changes

mcp-auth-oauth is a skill published in the GitHub repository latestaiagents/agent-skills (5 stars, last pushed 4mo ago), licensed MIT. It adds 108 tokens to every session and 1,647 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.