Sign in

MCP Server

Connect Steady to AI assistants

Steady's MCP server lets AI assistants like Claude interact directly with your Steady data — completing check-ins, pulling team digests, and surfacing insights — without leaving the tool you're already working in.

MCP (Model Context Protocol) is an open standard that connects AI assistants to the tools and data they need. With Steady's MCP server, your assistant can read and write Steady data alongside every other service you've connected.

What you can do

The MCP server covers all the core data and workflows in Steady:

  • Reading data from your Daily Digest, Smart Check-ins, Goal Stories, Goal Story updates, Echoes, activities, Insights, and absences.
  • Creating Check-ins, Goal Stories, and Goal Story updates.

Example use cases

  • Pull your team's check-ins from Steady, cross-reference blockers against Linear issues, and post a summary to Slack — all in one conversation.
  • Review your calendar and emails, check what your team reported in Steady, and draft a daily briefing.
  • Read a Slack thread, pull relevant Goal Story updates from Steady, and compose a stakeholder email in Gmail.
  • Push Steady context to other tools, like posting a digest to a Slack channel.

Server URL

https://app.steady.space/mcp

How to get started

You'll need an AI assistant that supports MCP — Claude, ChatGPT, Gemini, Copilot, and others all work.

Add Steady as an MCP server in your AI assistant

The exact steps depend on which assistant you use. Here's how it works in Claude:

Claude Desktop

  • Go to Settings
  • Select Connectors
  • Select "Add custom connector"
    • Name the connector "Steady"
    • Enter the server URL: https://app.steady.space/mcp
    • Select "Add"
    • Select "Connect" and sign into your Steady account

Claude Code

  1. Open Claude Code in your terminal.
  2. Run claude mcp add --transport http steady https://app.steady.space/mcp --scope user      
  3. This adds Steady as an MCP server available across all your projects. (Use --scope project to limit it to the current repo, or omit --scope for the current project only.)   

Then authenticate:

  1. Run /mcp   and select the Steady server.
  2. Choose "Authenticate" (or "Connect"). Claude Code opens your browser to the Steady login page.
  3. Sign in to Steady and approve the access request.
  4. Return to Claude Code. You should see: Authentication successful. Connected to steady.
  5. Run /mcp  again to verify Steady shows as connected.

Detailed guides for Claude and other apps:

(Optional) Connect your other services

Add MCP servers for Slack, Linear, Google Calendar, and any other tools you use. Each service has its own setup — most take under a minute.

Start using it

Ask your assistant to check you in, get your team digest, or combine Steady data with other services in a single conversation.

Demo

FAQs

Do I need a specific AI assistant? Any assistant that supports MCP works with Steady.

Is my data secure? MCP connections use OAuth 2.1 authentication. Your Steady data is only accessible to assistants you've explicitly authorized, and the same permissions you've set up in Steady apply.

Can my whole team use this? Yes. Each team member sets up their own MCP connection with their own Steady credentials. They'll only see the data they have access to — same permissions model as Steady itself.