Automation · 1 skill

n8n MCP and Claude skills

Read workflows and executions from n8n, and report failed runs, workflows that stopped running and error rates, with the node that broke each time.

Connect n8n to your agent (MCP)

n8n MCP server, built into n8n. An instance owner or admin turns it on under Settings > Instance-level MCP, and the server URL ends in /mcp-server/http. In Claude Code: claude mcp add --transport http n8n https://<your-n8n-domain>/mcp-server/http, then /mcp to sign in with OAuth, or add --header "Authorization: Bearer <token>" with the personal MCP token from the same settings page. In Codex, codex mcp add n8n --url https://<your-n8n-domain>/mcp-server/http and codex mcp login n8n; for Cursor, use the one-click setup in n8n's Connect a client dialog (docs). n8n is also in Claude's connector directory. Clients see only the workflows you enable for MCP (search returns previews of the rest), and with the right grants they can run, build and edit workflows.
Public API with an API key. Create one under Settings > n8n API, store it in N8N_API_KEY with N8N_BASE_URL, and send it in the X-N8N-API-KEY header. The API isn't available during n8n Cloud's free trial.

API key scopes are an Enterprise feature: there, give the key read access to workflows and executions only; on other plans a key has full access to your account. When you connect an MCP client for reporting, grant it read access without run or edit rights.

1 n8n skill

n8n failed executions report

Every morning: n8n workflows with failed executions since yesterday, how many times each failed, the error message and the node it stopped on.

DevOpsEvery day at 8:00Report card

What agents do with n8n

n8n's public API (https://<your-instance>/api/v1) lists workflows and executions. GET /workflows?active=true returns workflows with name, active, updatedAt and tags, and GET /executions?status=error&workflowId=...&limit=100 returns runs with status (success, error, waiting and others), mode, startedAt and stoppedAt; add includeData=true to get the run data that shows which node failed and its error message. Results page with cursor.

Automation owners schedule a morning report of failed executions in the last 24 hours grouped by workflow, workflows that were deactivated or haven't run when they should have, a weekly error rate per workflow, and a check on executions stuck waiting.

n8n.io

Teams that use these

Your n8n numbers, on one board with everything else

Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.

Join the open beta Browse all skills

Free during the open beta. No card.