Build on Meteor
From your first key to production: authentication, running Mets, streaming, CRM, webhooks and error handling.
How to build an AI agent
What an agent is, how it differs from a flow-based chatbot, and the five steps to having one answering on WhatsApp with access to your business data.
Authentication
Create your API key, understand scopes, and keep the key safe.
Run Mets (Runs)
Run your workspace's Met from the API, with optional memory, and read the result.
Third-party OAuth apps
Connect your app with user authorization, PKCE S256 and rotating tokens, without asking for API keys.
Streaming (SSE)
Receive the run's events live, one event at a time, with Server-Sent Events.
Tasks
Repeatable procedures a Met runs step by step, with pauses for a person to decide.
Functions and flows
Make your Met call your own backend — a Function with an HTTP handler or a flow, linked to the Met.
Automations
The *when* of your workspace: wire an event to an action and let Meteor fire it without anyone calling the API.
Mets and their tools
Create and adjust Mets, see which tools each one gets, and control which collections it can reach.
Skills
Give a Met new capabilities — activate them from the catalog, store their credentials and link them.
Images
Generate, edit and animate images through Meteor — same key, debiting Energy, no external provider.
Workspace variables
Store workspace values and secrets — like the token for an HTTP Function — and discover your CRM's fields.
Data: collections and items
Store structured data in collections, create it, search it by text and keep control of it.
MCP integrations
Register your own HTTP MCP server and run its tools from your Mets.
Events
Listen to your workspace's events live over SSE, or read them from the 30-day event log.
CRM and contacts
Create contacts, send them WhatsApp messages, tag them and walk your whole base.
Inbound webhooks
Register endpoints that trigger tasks or emit events, with HMAC signing.
Errors, idempotency and pagination
Typed errors, safe retries with idempotency, and how to paginate.
Limits and quotas
Per-key rate limits, rate limit headers, monthly quota, and how the SDKs retry on their own.
Templates (Snapshots)
Discover and install complete implementations — Mets, flows, collections, integrations — into a workspace with one API key.
CLI (met)
Run Mets, listen to events and forward them to your local app, from the terminal.
Workspace operations
Usage and billing, files, conversations, reminders, autopilot and agent groups — what you look up once you are in production.
Versioning and deprecation
Which changes can happen without notice, how an endpoint that will be retired is announced, and which headers tell you.
Account provisioning
Create Met accounts from your own product and top them up with energy, using a partner API key.
Recipes
End-to-end builds, from the first line to production. Each one is copy-paste runnable as it stands.
An agent that answers WhatsApp
From an incoming WhatsApp message to a Met's reply, with your own logic in the middle and a handoff to a person when it's needed.
Sync your system into collections
Mirror a catalog, an inventory or a price list inside Meteor so your Mets can read it without you writing an endpoint per question.
Receive signed events in your backend
An endpoint that receives your workspace's events, verifies the signature and survives retries without processing anything twice.
Run a Met from GitHub Actions
Put a Met inside your pipeline — summarize a deploy, review a change, run a task — with the exit code reflecting the result.
Show a run live in your UI
Relay the Met's stream from your backend to the browser, word by word, without exposing the key and without losing the result if the stream drops.
A task that asks for human approval
Let a Met do the work and stop to ask permission before the step that can't be undone, with the decision coming in from your own system.