Auth Setup logo

Auth Setup

Organization
butterbase-ai
auth-setup

Use when configuring OAuth providers (Google/GitHub/Apple/X/etc.), setting up post-login auth hooks, tuning JWT lifetimes, or generating service API keys

Overview

Publisherbutterbase-ai
Repositorybutterbase-skills
Skill nameauth-setup
Stars
535
Forks
42
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by butterbase-ai on GitHub. Read the source before you install it.

Installation

Install the Auth Setup AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/butterbase-ai/butterbase-skills/tree/main/skills/auth-setup
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/butterbase-ai/butterbase-skills.git /tmp/butterbase-skills
mkdir -p .claude/skills
cp -r /tmp/butterbase-skills/skills/auth-setup .claude/skills/auth-setup
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Auth Setup in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Auth Setup on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Auth Setup is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Butterbase Auth Setup

Two umbrella tools cover end-user authentication:

  • manage_oauth — provider configuration (Google, GitHub, Apple, X, custom)
  • manage_auth_config — auth hooks, JWT lifetimes, service key generation

For broad app build-out, see also butterbase-skills:build-app. This skill is the deep dive.


1. The role model

Every request runs under one of three database roles:

Auth headerRolecurrent_user_id()RLS
nonebutterbase_anonNULLenforced; default deny
End-user JWT (issued by manage_oauth or email login)butterbase_useruser UUIDenforced
Service key (bb_sk_*)butterbase_serviceNULLbypassed

Auth is what transforms a request into the right role. RLS is what filters the data. Both must be configured.


2. Configure an OAuth provider

js
manage_oauth({
  app_id: "app_abc123",
  action: "configure",
  provider: "google",
  client_id: "123456789.apps.googleusercontent.com",
  client_secret: "GOCSPX-...",
  redirect_uris: ["https://api.butterbase.ai/auth/app_abc123/oauth/google/callback"]
  // scopes / authorization_url / token_url / userinfo_url / provider_metadata are auto-filled for built-in providers
})

Built-in providers (URLs and scopes pre-filled): google, github, discord, facebook, linkedin, microsoft, apple, x.

Custom providers: pass authorization_url, token_url, userinfo_url, and scopes explicitly.

Redirect URI format

https://api.butterbase.ai/auth/{app_id}/oauth/{provider}/callback

Register this exact URI in the provider's developer console. Mismatch is the most common reason OAuth flows fail.

Provider quirks

ProviderQuirk
appleRequires provider_metadata: { teamId, keyId, privateKey }. Apple only returns the user's name on first auth and uses POST callback (handled automatically).
xDoes not return email. Butterbase synthesises {username}@users.noreply.x.local for the user record.
facebookDefault scopes email, public_profile.
googleStandard.
githubStandard.

List, update, delete

js
manage_oauth({ app_id, action: "get" })                          // list all providers (secrets redacted)
manage_oauth({ app_id, action: "get", provider: "google" })      // single provider
manage_oauth({ app_id, action: "update", provider: "google", client_secret: "new-secret" })
manage_oauth({ app_id, action: "delete", provider: "google" })   // disables future logins; existing sessions valid until expiry

Frontend flow

GET https://api.butterbase.ai/auth/{app_id}/oauth/{provider}?redirect_to=https://yourapp.com/auth/callback

User signs in at the provider, gets bounced back to redirect_to with access_token and refresh_token as query params. The Butterbase SDK wraps this:

ts
await client.auth.signInWithOAuth({ provider: "google" });
const { user, accessToken } = await client.auth.getSession();

3. Tune JWT lifetimes

js
manage_auth_config({
  app_id: "app_abc123",
  action: "update_jwt",
  accessTokenTtl: "15m",       // formats: "15m", "1h", "2h", "1d"
  refreshTokenTtlDays: 30      // integer days
})

Defaults: 15-minute access tokens, 7-day refresh tokens.

Use caseaccessTokenTtlrefreshTokenTtlDays
High-security (banking, admin)5m–15m1–7
Standard SaaS15m (default)30
Low-friction consumer apps1h90

Important: changes apply only to new tokens. Active tokens keep their original expiration — there is no global revoke. Treat TTL changes as forward-looking only.


4. Auth hooks (run code after every login)

A post-auth function is a deployed Butterbase function invoked fire-and-forget after every successful auth event (OAuth login, email login, email signup).

Wire it up

js
// 1. Deploy the function first (see butterbase-skills:function-dev)
deploy_function({
  app_id: "app_abc123",
  name: "after-auth",
  code: postAuthHandlerCode,
  trigger: { type: "http", config: { auth: "none" } }
})

// 2. Register it as the auth hook
manage_auth_config({
  app_id: "app_abc123",
  action: "configure_auth_hook",
  post_auth_function: "after-auth"
})

// To remove the hook later: pass post_auth_function: null

The function must already exist when you configure the hook.

Payload shape

The function receives a POST with this body:

json
{
  "event": "oauth_login | login | signup",
  "user": {
    "id": "uuid",
    "email": "...",
    "provider": "google | github | email | ...",
    "display_name": "...",
    "avatar_url": "..."
  },
  "isNewUser": true,
  "provider": "google"
}

The function runs as butterbase_service (RLS bypassed, ctx.user is null). Use body.user.id to know who just logged in.

Common uses

ts
// after-auth/index.ts
export async function handler(req, ctx) {
  const { user, isNewUser, event } = await req.json();

  if (isNewUser) {
    // 1. Create profile row
    await ctx.db.query(
      "INSERT INTO profiles (user_id, display_name) VALUES ($1, $2) ON CONFLICT DO NOTHING",
      [user.id, user.display_name]
    );

    // 2. Send welcome email (via env-stored API key)
    ctx.waitUntil(sendWelcomeEmail(ctx.env.RESEND_API_KEY, user.email));
  }

  // 3. Audit log on every login
  await ctx.db.query(
    "INSERT INTO login_log (user_id, event, provider) VALUES ($1, $2, $3)",
    [user.id, event, ctx.user ?? null]
  );

  return new Response("ok", { status: 200 });
}

Auth hooks are fire-and-forget. Don't return data the user needs — they won't see it. Use them for side effects only.


5. Service keys (bb_sk_*)

Service keys grant full access to all your apps and bypass RLS. Treat them like passwords.

Generate

js
manage_auth_config({
  action: "generate_service_key",
  name: "CI/CD pipeline"
})
// → { key: "bb_sk_a1b2c3...", key_id, prefix, name, created_at }

The full key is returned once. Store it immediately in your secret manager — you cannot retrieve it again. If you lose it, generate a new one and revoke the old.

List & revoke

js
manage_api_keys({ action: "list" })
manage_api_keys({ action: "revoke", key_id: "uuid-..." })

list returns metadata only (prefix, name, last_used_at), never the secret. revoke is immediate and irreversible.

Rotation workflow

  1. manage_auth_config (generate_service_key) — create the new key.
  2. Update CI/CD, MCP config, scripts to use the new key.
  3. Verify with a smoke test (e.g. manage_app list).
  4. manage_api_keys (revoke) — kill the old key.

Do steps 1–3 before step 4 to avoid downtime.


6. Anti-patterns

Don'tDo
Hardcode bb_sk_* keys in client code or commit them to gitStore in env vars / secret manager
Reuse one OAuth app between dev, staging, prodSeparate OAuth apps per environment, with their own redirect URIs
Use a service key from frontend code "for convenience"Frontends use end-user JWTs; service keys are server-only
Increase accessTokenTtl to "fix" frequent re-authUse the refresh token; SDK handles this automatically
Forget that manage_oauth delete only stops new loginsExisting sessions remain valid until they expire — rotate JWT keys via support if you need a hard kill
Put critical logic in the auth hookHooks are fire-and-forget. Errors don't surface to the user. Keep them to side effects.
Log the full service key in audit / debug outputLog only the prefix (bb_sk_a1b2c3) — secrets must never appear in logs

7. Quick reference

TaskTool
Add Google OAuthmanage_oauth (configure)
List OAuth providersmanage_oauth (get)
Set post-login hookmanage_auth_config (configure_auth_hook)
Change JWT lifetimesmanage_auth_config (update_jwt)
Create service keymanage_auth_config (generate_service_key)
List service keysmanage_api_keys (list)
Revoke service keymanage_api_keys (revoke)

If a docs/butterbase/00-state.md exists in the working directory, prefer invoking via /butterbase-skills:journey-auth so the journey orchestrator stays in sync.

Frequently asked questions

What does the Auth Setup AI skill do?

Use when configuring OAuth providers (Google/GitHub/Apple/X/etc.), setting up post-login auth hooks, tuning JWT lifetimes, or generating service API keys

Why use Auth Setup on TypingMind?

Because you install it once and use it with any model. Auth Setup is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Auth Setup in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/butterbase-ai/butterbase-skills/tree/main/skills/auth-setup. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Auth Setup?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Auth Setup?

As many as you like. As long as a model supports skills, you can use Auth Setup with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Auth Setup AI skill free?

Yes. It is published on GitHub by butterbase-ai under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇