Steady MCP
Enables submitting team check-ins to Steady (app.steady.space) by automating the web form through login, team discovery, and check-in submission with previous/next/blockers fields.
README
Steady MCP (local)
This is a local MCP server that submits Steady check-ins by driving Steady’s web form (there is no official Steady API).
It’s built for the workflow:
You tell the AI: team + check-in fields (previous / next / blockers) → AI calls the MCP tool → Steady check-in is submitted.
What this supports
- Email + password login (two-step flow from
/signin) viasteady_login - Cookie-based auth via
steady_set_cookies(fallback for non-standard auth flows) - Team discovery via
steady_list_teams - Submitting a check-in for one team via
steady_submit_checkin
How it works (important)
Steady rotates the _sthr_session cookie when you load the daily edit page (/check-ins/YYYY-MM-DD/edit).
So this server uses a curl cookie jar (cookiejar.txt) to preserve the updated session cookie between the GET and POST.
Requirements
- Node.js 18+
- Cursor (or another MCP client) with support for running local MCP servers over stdio
- Access to Steady web app (
https://app.steady.space)
Install
cd tools/steady-mcp
npm install
Configure Cursor (MCP)
Add a server entry to your Cursor MCP config.
Where is mcp.json?
Common locations:
- macOS / Linux:
~/.cursor/mcp.json - Windows:
%USERPROFILE%\\.cursor\\mcp.json
If you don’t see it, use Cursor’s UI settings for MCP servers (recommended), or search for mcp.json on your machine.
Quickstart: macOS (recommended)
- Install Node.js 18+
- If you use Homebrew:
brew install node
node -v
- Install dependencies
cd <ABS_PATH_TO_REPO>/tools/steady-mcp
npm install
-
Add
steady-mcpto~/.cursor/mcp.json(example below) -
Restart Cursor (quit + reopen)
-
In Cursor, run:
steady_loginsteady_submit_checkin
Quickstart: Windows
- Install Node.js 18+
- If you use
winget:
winget install OpenJS.NodeJS.LTS
node -v
- Install dependencies
cd <ABS_PATH_TO_REPO>\tools\steady-mcp
npm install
-
Add
steady-mcpto%USERPROFILE%\.cursor\mcp.json -
Restart Cursor (fully quit + reopen)
-
In Cursor, run:
steady_loginsteady_submit_checkin
Example mcp.json snippet
Replace
<ABS_PATH_TO_REPO>and your email.
{
"mcpServers": {
"steady-mcp": {
"command": "node",
"args": [
"<ABS_PATH_TO_REPO>/tools/steady-mcp/src/index.js"
],
"env": {
"STEADY_BASE_URL": "https://app.steady.space",
"STEADY_EMAIL": "you@company.com",
"STEADY_PASSWORD": "<YOUR_PASSWORD>",
"STEADY_MCP_DEBUG": "0"
}
}
}
}
Windows path tip (important)
In JSON, Windows backslashes require escaping. Easiest option: use forward slashes in the args path:
{
"mcpServers": {
"steady-mcp": {
"command": "node",
"args": [
"C:/Users/<YOU>/path/to/repo/tools/steady-mcp/src/index.js"
],
"env": {
"STEADY_BASE_URL": "https://app.steady.space",
"STEADY_EMAIL": "you@company.com",
"STEADY_PASSWORD": "<YOUR_PASSWORD>"
}
}
}
}
Security note
STEADY_PASSWORDinmcp.jsonis the simplest setup, but it’s not ideal for security.- Prefer
STEADY_PASSWORD_COMMAND(password retrieved at runtime) whenever possible.
Credentials / Secrets (cross-platform)
The server reads credentials from environment variables provided by your MCP client.
Option A (simple): plaintext env password
STEADY_EMAILSTEADY_PASSWORD
Option B (recommended): password command
STEADY_EMAILSTEADY_PASSWORD_COMMAND— command that prints the password to stdout (no prompts)
Examples:
- macOS Keychain (recommended on macOS)
- Store once:
security add-generic-password -a "you@company.com" -s "steady-mcp" -w "<YOUR_PASSWORD>" -U
- In MCP env:
STEADY_PASSWORD_COMMAND=security find-generic-password -w -s steady-mcp -a you@company.com
- Windows: use a password manager CLI (e.g. 1Password CLI / Bitwarden CLI) or a small local script that prints the password.
- Any OS: use your password manager’s CLI, or a small script you keep outside git.
Option C (macOS-only fallback): Keychain lookup without a command
If you have a Keychain entry, the server can read it automatically on macOS:
STEADY_EMAIL=you@company.com- optionally
STEADY_KEYCHAIN_SERVICE=steady-mcp - optionally
STEADY_KEYCHAIN_ACCOUNT=you@company.com
Authentication options
1) Login automation (email → password)
Use:
steady_login
This will log in and write:
cookies.txt(Cookie header string)cookiejar.txt(curl jar; used internally to handle session rotation)
2) Cookie-based auth (fallback)
If login automation doesn’t work for your Steady account, you can set cookies from your browser:
- In Steady (browser): DevTools → Application/Storage → Cookies →
https://app.steady.space - Copy these cookies:
_sthr_sessionremember_user_token(if present)
- Build a Cookie header string:
_sthr_session=...; remember_user_token=...
- Call:
steady_set_cookieswith{ "cookies": "_sthr_session=...; remember_user_token=..." }
File locations (where cookies are stored)
Defaults (can be overridden via env):
- macOS
- cookies:
~/Library/Application Support/steady-mcp/cookies.txt - jar:
~/Library/Application Support/steady-mcp/cookiejar.txt
- cookies:
- Windows
- cookies:
%APPDATA%\\steady-mcp\\cookies.txt - jar:
%APPDATA%\\steady-mcp\\cookiejar.txt
- cookies:
- Linux
- cookies:
$XDG_CONFIG_HOME/steady-mcp/cookies.txt(or~/.config/steady-mcp/cookies.txt) - jar:
$XDG_CONFIG_HOME/steady-mcp/cookiejar.txt(or~/.config/steady-mcp/cookiejar.txt)
- cookies:
Overrides:
STEADY_COOKIES_PATHSTEADY_COOKIE_JAR_PATH
Usage (day-to-day)
Step 0: Restart Cursor
After adding/updating MCP config, fully restart Cursor (quit + reopen).
Step 1: Login (refresh cookies)
Call:
steady_login
Then verify:
steady_ping→ should return{ "ok": true }
Step 2: List team names (optional)
Call:
steady_list_teams
Step 3: Submit today’s check-in for one team
Call:
steady_submit_checkin
Example:
{
"team": "Everest AI",
"previous": "Wrapped up CI fixes and reviewed PRs.",
"text": "Next: finalize the MCP schema changes and update docs.",
"blockers": "No blockers.",
"mood": "calm"
}
Success is typically:
status_code: 302
Tools (API surface)
steady_login: log in and save cookies locallysteady_set_cookies: save browser cookies manuallysteady_ping: validate authsteady_list_teams: list team names/ids from the daily edit pagesteady_submit_checkin: submit for one team
Troubleshooting
Start here: docs/TROUBLESHOOTING.md
Common issues:
-
Tool not found in Cursor
- fully restart Cursor
- verify the
mcp.jsonentry path is correct and absolute - verify
npm installwas run intools/steady-mcp
-
HTTP 422
- usually means the submission was rejected (already checked in / stale session / insufficient permission)
- run
steady_loginagain and retry - confirm the team is actually pending on Steady’s daily page
-
Need debug
- set
STEADY_MCP_DEBUG=1(warning: debug output may include sensitive cookies)
- set
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.