AgentReady exposes a Model Context Protocol (MCP) endpoint so you can drive it from the AI agent you already use. Point Claude Code, Claude Desktop, Cursor, or anything that speaks MCP at your workspace, and it reads the same record you see in the dashboard: your profile, your tracked questions, the answers with their evidence, the fix queue, and the verification ledger.
The honest hook is simple. Your agent reads the same evidence you do, and verification never takes its word for it.
Rolling out. The agent endpoint ships with the current release wave. This page describes the contract so you can plan for it, but the live endpoint URL and your workspace token appear in your dashboard settings when the surface is switched on for your workspace. Until then, treat the addresses below as the shape of what you will paste, not a URL to hit today.
What it is
Two things, issued from your dashboard:
- An MCP endpoint hosted at
https://app.joinagentready.com/api/…. The exact path is shown in your dashboard settings when available. It speaks streamable HTTP, the transport every current MCP client supports. - A workspace-scoped token. It authenticates the endpoint and scopes every call to one workspace, yours, and nothing else. You create it, see it, and revoke it from settings. It is shown once at creation and hashed at rest, so a leaked token is inert the moment you revoke it.
You paste both into your client once. From then on your agent talks to AgentReady over that connection.
Set it up in your client
No client is privileged. Use whichever you like. In every block, replace the bracketed placeholders with the endpoint and token from your dashboard settings.
Claude Code
claude mcp add --transport http agentready \
"https://app.joinagentready.com/api/<mcp-endpoint-from-settings>" \
--header "Authorization: Bearer <your-workspace-token>"
Claude Desktop
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"agentready": {
"type": "http",
"url": "https://app.joinagentready.com/api/<mcp-endpoint-from-settings>",
"headers": {
"Authorization": "Bearer <your-workspace-token>"
}
}
}
}
Cursor
Add this to .cursor/mcp.json in your project (or your global Cursor MCP config):
{
"mcpServers": {
"agentready": {
"url": "https://app.joinagentready.com/api/<mcp-endpoint-from-settings>",
"headers": {
"Authorization": "Bearer <your-workspace-token>"
}
}
}
}
Any other MCP client
Point a generic streamable-HTTP MCP client at the same values:
Transport: Streamable HTTP
URL: https://app.joinagentready.com/api/<mcp-endpoint-from-settings>
Header: Authorization: Bearer <your-workspace-token>
What your agent can do
Read the record. Your agent can pull your business profile, your tracked question set, your latest runs and answers (each carrying the evidence it was scored from), your fix queue with its jobs and skip reasons, and your verification ledger. This is the same data the dashboard shows, read straight from the system of record.
Confirm profile facts. Your agent can confirm the business facts the profile is built on, your categories, your policies, the details it works from. Each confirmation is recorded with owner-via-assistant provenance and is revertible, so the record always shows who confirmed what and through which channel.
Apply fixes on your own property. For changes that live on a surface you control, your agent can apply the fix using the structured application contract that ships with each artifact: the target, the exact patch, and the acceptance criteria. After it applies, AgentReady re-crawls your live property and judges the result itself. Verify-by-recrawl closes the loop without trusting the agent's claim that the change landed.
What it cannot do
The write path is governed on purpose, and the limits are structural, not a setting you can flip.
- It cannot publish or apply to surfaces AgentReady manages without your in-app approval. Approval stays in the app and is never delegated to the conversation. An agent can prepare and propose; only you, in AgentReady, release anything outward.
- It cannot act without leaving a record. Every agent action appends to the ledger with its actor and channel. Nothing rewrites history, so you can always see what the agent did and undo it.
- It cannot outlive a revoke. Your token is visible and revocable from settings, and killing it cuts the connection immediately. Tools that must not exist on this surface are simply absent from it, not hidden behind a permission you might grant by accident.
That is the whole point of the hook: your agent reads the same evidence you do, it can do real work on your own property, and verification never takes its word for it.
One surface, or two?
If you have used AgentReady on a Shopify store, you may have met the storefront MCP already. They are different surfaces with one job each, and you can keep both:
- The storefront MCP is shopper-facing. It gives an assistant read-only access to your live products, policies, and brand info so it can research and answer questions about your store. It sits alongside Shopify's native shopper MCP and is not how you operate AgentReady.
- The agent endpoint on this page is owner-facing. It reads your AgentReady record and, with your approval rails in place, helps you operate the app: confirm facts, walk the fix queue, and apply on your own property.
When you connect your agent to operate AgentReady, the endpoint on this page is the one you want.