Guidewheel

Active

An AI agent for Guidewheel. It works in a scene with a live connection to Guidewheel. It asks before it writes, and every run is traced.

Connect Daslab scenes to Guidewheel. Pin machines and downtime issues as assets, read run/idle/down states, uptime, energy, and production actuals — and write back issues, production entries, and scrap to close the loop with your ERP or CMMS.

Data & Databases Custom 4 asset types

What you can connect

Add these to your scene and AI gets access.

Guidewheel Account Guidewheel Accounts

A Guidewheel FactoryOps tenant, connected via base URL and API key.

NameTenant

Machine Machines

A sensored machine, line, or cell in Guidewheel.

NamePlantState

Downtime Issue Downtime Issues

A downtime event or anomaly detected by Guidewheel's alerting.

TitleStatusReason

Guide Guides

Guidewheel setup guides, troubleshooting, and reference — pin one to read it in-scene, or ask the agent.

TitleSummary

Connect Guidewheel over MCP

Daslab is itself an MCP server.

Connect Guidewheel in a scene, then pin that scene's URL in Claude Code, Cursor, or any MCP client. The agent there gets your Guidewheel tools, and your approval gate and trace travel with them.

https://daslab.run/mcp/<your-world>/<your-scene>

One scene can hold Guidewheel and everything else you connect, so a single endpoint carries them all. Set up a client.

Use cases

Pin a machine and see its live run/idle/down state
Turn a downtime issue into an ERP/CMMS action and write the reference back
Reconcile production actuals and energy per shift into reports

Knowledge base

Getting Guidewheel API access api-access
No developer portal — request the tenant base URL, company-scoped API key, and sandbox from Guidewheel; auth, rate limits, and environments.
Data model & endpoint map data-model
What the CT-sensor signal can tell you, the /api/v1 endpoints and verbs, and the gaps to plan around (no webhooks, no computed OEE).
Closing the ERP/CMMS loop erp-cmms-loop
The two integration patterns — planned orders in / actuals out, and downtime issue → work order with status write-back — as agent flows.

Ships with the integration — agents read these with docs_read, before the account is connected. Knowledge bases →