OpenIT puts an AI agent on your team. It handles tickets, provisions access, writes integrations, and learns everything you teach it — so you stop answering the same question twice.
Every other ITSM is configured through clicky admin screens and proprietary workflow builders. OpenIT inverts it — you describe what you want in Claude Code; the result is plain files you can read, edit, and take with you anywhere.
Describe your processes in plain English. Create skills, automation scripts, integrations — all in the terminal, all version-controlled.
When someone asks in Slack or files a ticket, Claude reads the context, applies your automations, or escalates with full context.
Every answer you give becomes a saved automation. First-time questions become second-time resolutions — the queue gets smarter every day.
Employees ask in Slack. Claude answers, or escalates with full context. Socket Mode — no public endpoint needed.
Claude uses the same tools you do — brew, gh, aws, gcloud, kubectl. No middleware. If you can run it in a terminal, Claude can use it.
One-click install from the Tools station. Claude connects to MCP servers the same way you do — direct, no vendor lock-in.
Every skill, script, and KB article is a plain file on your machine. Read them. Edit them. Check them into git. Take them with you.
TypeScript, Python, Bash — Claude writes and runs scripts when a skill isn't enough. Scripts become skills. Skills become muscle memory.
Every fix generates a KB article in plain English. Your team's knowledge compounds automatically — no one has to write docs.
Fork it. Ship it. Make it yours. macOS and Windows today, Linux coming — contributions welcome.
The first time someone asks something, Claude escalates to you. After you answer, it saves the automation. Two weeks in, the queue handles itself — and you spend your days on the work that actually matters.