How do I monitor n8n AI workflows?

Install ClawMetry (pip install clawmetry). The daemon auto-detects ~/.n8n (or N8N_USER_FOLDER if set) and renders every workflow execution, AI node call, and per-node token spend in one dashboard at http://localhost:8900.

Does ClawMetry work with n8n's LangChain nodes and OpenAI nodes?

Yes. Both carry usage metadata on their execution output; ClawMetry parses the LangChain shape (usage_metadata) and the OpenAI shape (prompt_tokens + completion_tokens) and normalises them so per-node cost is directly comparable.

Can I see cost per n8n workflow, per hour?

Yes. The Cost tab groups by workflow id and shows per-workflow, per-hour token spend. Set an alert on a per-workflow budget and the approval queue pauses subsequent executions.

Does ClawMetry catch a scheduled n8n workflow that loops overnight?

Yes. n8n's cron trigger fires executions on schedule; ClawMetry watches execution frequency and per-execution token growth. A workflow whose token spend accelerates hour-over-hour is flagged, and the kill switch (Pro) can pause future scheduled executions.

How much does ClawMetry cost for n8n?

Free forever on your own machine. Cloud starts at $9/node/mo (Starter) or $19/node/mo (Pro).

Observability and cost tracking for n8n AI workflows

See every n8n AI workflow execution, node call, and token as it happens. ClawMetry reads the n8n data directory under ~/.n8n and renders workflow runs, per-node cost, and per-provider token spend in one dashboard.

Supported runtime Reads n8n's data folder (SQLite executions store + node logs) Read-only Free trial

What ClawMetry sees on n8n

ClawMetry's n8n adapter opens the store n8n already writes (~/.n8n (overrideable via N8N_USER_FOLDER)) in read-only mode and surfaces:

Sessions: started at, last activity, model, message count, per-session token totals.
Live activity: every tool call as it lands, with args and result. Turn-level tokens split.
Per-turn tokens: Per-node usage extracted from n8n's AI node executions (LangChain nodes carry the LangChain usage shape; OpenAI Chat nodes carry the OpenAI shape).
Cost per session, per model, per hour: cache-aware where the model supports caching.
Per-workflow-node cost attribution (which AI node in which workflow spent how many tokens): the n8n-specific signal ClawMetry surfaces that generic dashboards do not.
Approval and kill switch: watch for token-maxing signals (message repeat, no-progress heuristic, cost-per-hour threshold), pause or kill before the next call.

Source: the adapter is at clawmetry_pro/adapters/n8n.py. Every parse rule and every SQL SELECT is in that file.

Enable it in about two minutes

1. Install ClawMetry

Pick the line that matches your OS. Windows users: paste the install.cmd line, not the shell one.

# macOS or Linux
curl -fsSL https://clawmetry.com/install.sh | bash

# Windows (cmd.exe or PowerShell)
curl -fsSL https://clawmetry.com/install.cmd -o install.cmd && install.cmd

2. Run the daemon

clawmetry

The daemon auto-detects n8n's local store and starts reading immediately.

3. Open the dashboard

Open http://localhost:8900. Switch the runtime chip to n8n.

Start the free trial →

The one thing this catches that n8n alone does not

Left alone, n8n's failure mode is a scheduled n8n workflow that quietly runs an OpenAI node in a loop through the weekend. ClawMetry watches for it: cost per hour on the n8n runtime, message-repeat detection, and per-session token-growth curves. Set a threshold, and the approval queue pauses before the next call. Kill switch (Pro) stops it.

All 19 runtimes ClawMetry observes

n8n is one of 19. Free on every plan: OpenClaw, NVIDIA NemoClaw. Paid on Cloud Pro or Self-Hosted Pro: the rest.

OpenClaw NVIDIA NemoClaw NanoClaw PicoClaw Claude Code Codex Cursor Aider Goose opencode Qwen Code Hermes Pi Deep Agents n8n (this page) Antigravity GitHub Copilot Grok QM