Every coding agent keeps its own session store, its own auth, its own todo list and its own idea of what a “project” is. Use two of them on the same codebase and you have two histories, two usage bills and no single place to look. And none of them tell you what was actually delivered — only what was spent.
I wrote last year about the observability gap between what AI consumes and what it produces. Tokens measure input. Nobody measures output. That gap is the reason most enterprises have deployed agents and almost none have them in production at scale.
XO Space is the thing we built to close it.
XO Space is the environment layer between your machine and your agents: a local server that sits in front of the agents you already have installed, plus a browser UI that reads what they do. It does not run models or replace the agents. It stitches them together and measures what they deliver.
curl -fsSL https://quirq.ai/install | sh
# open http://localhost:5002/space/One command. One screen. Every agent, every project, every session, every cost.
👀 Watch your agents work — see which agent is active in which project right now, what it touched, and what it’s been doing session by session. No more alt-tabbing between five terminal windows to figure out which agent is doing what.
🗂️ Your whole workspace in one place — every folder is a project. Files, git history, todos and live agent activity, browsable from one screen. The Files tab shows every project in the workspace, which agent is active in it, and last activity. The Timeline shows every project’s git history in parallel lanes — click any commit to open a 3D city map where building height is churn.
✅ Manage the work, not the chat — todos per project, kept up to date as agents check them off. You always know what’s done and what’s next without scrolling through a hundred-message chat log.
📡 Sessions across every agent — Claude Code, Codex, OpenClaw, Hermes, Antigravity, even Cursor’s telemetry. Session history from every agent that exposes it, side by side, regardless of which one you’re chatting with right now.
📈 Know what it cost — tokens, cost, per-model and per-tool breakdowns in one shape regardless of runtime. Not an estimate from a vendor dashboard. The actual cost of what each agent did, per project, per session.
The reason most AI deployments stall at 11% production adoption isn’t model quality. It’s accountability. Nobody can answer the question a CFO actually asks: what did the AI finish, what evidence supports it, and what did it cost?
XO Space answers that question out of the box. Every project has a cost ledger. Every session has a record. Every todo has a completion trail. The data is local, portable, and lives next to your code — not inside a vendor’s database.
That last property matters more than it sounds. When your observability data lives in the same repository as your code, it travels with the project. Fork the repo, move to a different machine, onboard a new teammate — the history comes with you. It’s not locked in a dashboard you lose access to when you cancel a subscription.
| Agent | Chat | Sessions |
|---|---|---|
| Claude Code | ✅ | ✅ |
| Codex | ✅ | ✅ |
| OpenClaw | ✅ | ✅ |
| Hermes | ✅ | ✅ |
| Antigravity | ✅ | ✅ |
| Cursor | — | ✅ (read-only) |
Switch between them with one environment variable. Or run all of them at once — the Sessions tab shows telemetry from every agent that has it, regardless of which one is active.
Run this from the directory you want as your workspace:
curl -fsSL https://quirq.ai/install | shThe installer clones XO Space, creates a Python venv, and starts the server on localhost:5002. Open http://localhost:5002/space/ and you see every folder as a project, every agent that’s been active, every todo and every cost.
Machine-local state and logs live in ./.quirq/, next to your projects. The whole install is one folder you can move or delete. No account required. No data leaves your machine until you sign in.
I’ve been saying for a while that tokens are an input meter, not a measure of value. XO Space is the practical expression of that argument. It gives you the operational record — what was tried, what was accepted, what it cost — without making the model tell you about itself.
The next step is connecting that record to acceptance: marking work as delivered, linking it to the evidence trail, and letting the organization build a ledger of what its AI actually produces. That’s what we’re building toward with quirqs.
For now, the observability layer is live, it’s open source, and it runs on your machine in one command. If you’re running more than one agent on more than one project, you need a single place to see all of it.
This is that place.