The Developer API lets your own scripts and integrations read and update your CRM, and the MCP server lets Claude and other AI agents work with it too. Both use the same API keys, under the same permissions as the rest of your team.
Your API reference lives inside your organization at
https://<your-org>.guardenerai.com/developers/
Sync records with another system, automate a routine from a script, or let an AI agent work a queue. The API covers the records your team works with every day.
Read and update companies, people, leads, opportunities, tasks, cases and notes.
List the items waiting in a queue, then claim one to work on or release it back.
Attach files to records, so documents land next to the customer they belong to.
Read invoices from your built-in books. Accounting data is read-only in this version.
Connect Claude or another MCP-capable agent so it can search, read and update records for you.
Every organization has its own API reference with in-browser try-it, an OpenAPI spec and a Postman collection.
An organization admin creates keys under Settings, API keys. Send the key with
every request in the X-Api-Key header.
Each organization's reference lives at its own address:
https://<your-org>.guardenerai.com/developers/ It includes in-browser try-it, an OpenAPI spec download and a Postman collection. Authenticate with:
X-Api-Key: <your key> The MCP (Model Context Protocol) server lets MCP-capable AI agents work with your records, using the same API keys as the Developer API. No separate credential.
https://<your-org>.guardenerai.com/mcp (streamable HTTP)x-api-key and x-org-base-url: https://<your-org>.guardenerai.comFor example, with Claude Code:
claude mcp add guardenerai --transport http https://<your-org>.guardenerai.com/mcp \
--header "x-api-key: <your key>" \
--header "x-org-base-url: https://<your-org>.guardenerai.com" A key can never do more than a Member of the organization. Read-only keys cannot change anything.
Only an admin creates keys. Rotate them, set them to expire, or revoke one and access ends immediately.
Everything an AI agent does follows its key's permissions, read-only scope, usage limits and rate limits.
Want the wider picture? Read how we approach security.
An organization admin opens Settings, API keys, creates a key and copies it. It is shown once.
Go to /developers on your organization's address to browse endpoints and try them in the browser.
Send the key in the X-Api-Key header from your code, or add the MCP server to your AI agent.
Start free, create a key, and make your first call today. Have a question about an integration? We are happy to talk it through.