Agent-Ready Platform

Built for Agents

18 backend blocks. 203 skills. One API. Let your AI agents build with production-ready infrastructure.

18
Blocks
203
Skills
REST
API

How it works

Three steps for any agent

Same API your frontend uses. Now your agents can use it too.

01

Discover

Agent reads llms.txt or installs an MCP plugin to learn available blocks and skills.

02

Authenticate

Agent registers via AID protocol or uses your app-id and JWT. Two paths — admin-initiated or agent-initiated.

03

Execute

Agent calls skills — create users, upload files, manage data. 203 capabilities across 18 blocks.

Agent surfaces

Multiple ways for agents to integrate with 23blocks

MCP Plugins

Install 23blocks as Claude Code MCP servers. Each block is a standalone plugin with typed tools.

Claude CodeMCP Protocol18 Plugins

REST API

Standard HTTP endpoints for any agent or framework. JSON in, JSON out. Works with every LLM and agent runtime.

Any AgentHTTP/JSON203 Endpoints

llms.txt

Machine-readable product summary following the llms.txt standard. Agents can discover capabilities automatically.

Auto-Discoveryllmstxt.orgPlain Text
NEW — Agent Identity (AID)

Secure agent registration

Two paths for agents to authenticate with your backend. Based on RFC 8628 device authorization.

Admin-Initiated

Direct registration

1

Admin registers the agent with name, scopes, and Ed25519 public key

2

Agent receives JWT credentials and starts calling APIs immediately

Ed25519 KeysScoped AccessInstant Setup

Agent-Initiated

RFC 8628 device flow

1

Agent requests registration and receives a user code (XXXX-XXXX)

2

Agent displays the code — admin approves it in the dashboard

3

Agent polls for approval and receives JWT tokens automatically

RFC 862824hr Code TTLNo Shared Secrets

Get started in one command

Install the Auth block as a Claude Code MCP plugin

terminal
$ claude mcp add-json "23blocks-auth" '{"type":"stdio","command":"npx","args":["23blocks-auth-mcp"]}'

Frequently Asked Questions

Everything about agents and 23blocks

Open Source

Need multi-agent orchestration?

AI Maestro coordinates multiple AI agents with persistent memory, agent-to-agent messaging, and a peer mesh network. Open source under MIT license.

Your backend is agent-ready

Sign up, get your API keys, and let your agents start building.

Stay in the loop

Get product updates, engineering posts, and new block announcements delivered to your inbox.

No spam. Unsubscribe anytime. Privacy policy.