No ticket portal, no tier-one script, no chatbot pretending to be a person. A shared channel in Slack or Teams with engineers who can change the thing you are complaining about.
Support lives where your work lives. If you are a customer, you already have a channel — use it rather than email.
A shared channel between your workspace and ours, set up during onboarding. Engineers, your account lead, and the person who wrote the feature are all in it. Ask in the open — other people benefit.
External federation if Teams is your surface. Same people, same channel model, same response targets. Your admin approves the federation once.
Email hello@wildfire.business. It goes to a human in Copenhagen, not a queue.
Targets, not marketing. If we miss one, it shows up in the channel before you have to chase it.
| Severity | First response | What counts |
|---|---|---|
| Critical | 1 hour, 24/7 | Platform down, agents unreachable, or a suspected data-handling incident. Escalates to on-call immediately. |
| High | 4 business hours | A core workflow is broken for a team. Connector down, tier routing misbehaving, ledger not writing. |
| Normal | 1 business day | Something works but not well. A skill producing poor output, a connector scope question, a how-do-I. |
| Request | 1 week, with an answer | New connector, new feature, a change to the roadmap. The answer may be no, but it will be a real answer. |
Business hours are 09:00–17:00 CET, Monday to Friday. Critical is always covered.
The five steps, core concepts, and the admin settings people ask about most.
Usually an expired OAuth token or a scope your IT team narrowed. Both are visible in Settings.
Auth, endpoints, rate limits, and the MCP client configuration.
Article 14 exports, retention windows, sub-processor list and the residency table.
What you owe as a deployer and the twelve-point checklist.
The changelog, updated every Friday, including what broke.