/
Connect an AI assistant to your Cronium instance and build automations from plain language — it drafts events and workflows, creates them on your approval, runs them to verify, reads the results, fixes what's broken, and activates them.
Cronium ships a Model Context Protocol (MCP) server, so any MCP-capable AI app can build automations for you. You describe what you want; the assistant reads your Cronium's capabilities (tool types, your saved credentials, scheduling rules), drafts the events and workflow — filling in sensible defaults — and, once you approve, creates them. It can then run the automation, inspect the execution and logs, fix anything that failed, and activate it once it works — the whole loop without leaving the conversation.
For example, you can say:
“Create a workflow in Cronium: a SQL event that counts the users in the users database, then a Slack event that posts the count to #general at 8am every day.”The assistant creates a two-step workflow — a SQL query feeding a scheduled Slack message (0 8 * * *) — with Local execution and a 30s timeout filled in. Everything lands as a draft you review and activate.
The assistant reads your tool credentials, the available actions, and Cronium's scheduling rules so it drafts valid config — not guesses.
It can dry-run the plan (schemas, credentials, workflow graph) without changing anything, then show you the draft for approval.
Everything is created as a draft. With your approval the assistant can run it once, read the logs, fix failures, and activate the schedule when it's green.
There are two ways to connect, depending on which AI app you use.
Add a custom connector in Claude pointing at your instance:
https://your-cronium-host/api/mcpCronium is a full OAuth 2.1 authorization server for this connector, so you just click Connect, sign in to Cronium, and approve — no token to paste. (Any tier of the Claude apps can add a custom connector; a paid plan is needed for more than one.) Alternatively, authenticate with a bearer API token as a custom header. Your instance must be reachable over HTTPS.
For local clients, use the published stdio bridge @cronium/mcp-server — no install or clone needed, it runs via npx and forwards every request to your instance's /api/mcp endpoint. Set CRONIUM_BASE_URL to where your instance runs — a production deployment is http://localhost:3000 (or your domain); the examples below use the local development server's port 5001. Claude Desktop — claude_desktop_config.json:
{
"mcpServers": {
"cronium": {
"command": "npx",
"args": ["-y", "@cronium/mcp-server"],
"env": {
"CRONIUM_BASE_URL": "http://localhost:5001",
"CRONIUM_API_TOKEN": "<your-token>"
}
}
}
}Claude Code:
claude mcp add cronium \
--env CRONIUM_BASE_URL=http://localhost:5001 \
--env CRONIUM_API_TOKEN=<your-token> \
-- npx -y @cronium/mcp-serverget_capabilities (enums, defaults, scheduling guidance, available tool actions, and your real credential/server ids) and validate_plan (dry-run a draft without creating anything).list_events, get_event, get_event_logs, list_workflows, get_workflow, get_executions, get_execution — paginated summaries; env-var values and credential secrets are never returned.create_event, update_event, create_workflow, update_workflow, and create_workflow_bundle (several events and a workflow chaining them, in one step).activate_event, deactivate_event, delete_event, delete_workflow.run_event and run_workflow trigger a real run (works on drafts), then get_event_logs / get_execution show what happened.The easiest path is Settings → Connect AI in your instance: it shows your endpoint URL, mints an MCP-scoped token in one click, and fills both into ready-to-paste snippets for each client.
The remote connector can use OAuth (recommended) or a bearer API token; the local bridge uses an API token. You can also create tokens under Settings → API Tokens — every token needs an explicit scope and an expiry. Turn on “Limit to MCP” to make it least-privilege: such a token (and any OAuth connection) can only build, run, and manage events and workflows — it can't touch the rest of your account, so a leaked connector token is contained.
run_event / run_workflow execute your real scripts and tool actions — the assistant asks before triggering one, and deletes require confirmation too.source = "mcp" and logged, so you can always tell what an AI agent created.MCP is an open standard, so the same server works with other MCP-capable clients (e.g. ChatGPT's custom connectors on Business/Enterprise plans, and developer tools like Cursor and VS Code). Remote clients use the /api/mcp endpoint; clients that only support local servers can run the bundled stdio bridge, which forwards to the same endpoint.