Introducing CFactory: a control tower for your software factory
Three autonomous tools are powerful. Three you can see, correlate and steer from one place are a factory.
The Factory family already does something remarkable: it plans software (PFactory), builds it (AIFactory), and verifies it (TFactory) — autonomously, with humans in the loop only where it matters.
But there was a gap, and it was the obvious one. Each service runs on its own portal, its own workspace, its own status model. To answer a simple question — “where is this feature right now, and why is it stuck?” — you had to open three tabs and reconstruct the story by hand. The pipeline was real; the view of it wasn’t.
CFactory is that view.
A cockpit, not another dashboard
CFactory threads every unit of work across the three services into a single
WorkItem, keyed by its GitHub issue number:
plan → code → branch / PR → tests
One board shows where everything is — with gates, verdicts and PR links — fed live by REST, WebSocket and webhooks from the three services. No new agent logic; CFactory observes through the APIs that already exist.
And a copilot that actually knows the pipeline
On top of the board sits an agentic copilot (built on the Claude Agent SDK). Ask it “why is #182 stuck?” and it answers from real cross-service state — summarising the timeline, rolling up cost and latency, and flagging anomalies like runaway handback loops or cost spikes.
Crucially, it’s advise-and-confirm. The copilot can prepare an action — approve a gate, trigger a handoff, kick a stalled handback — but nothing is written until a human clicks. That’s the human-in-the-loop “Red Zone” guardrail the whole family is built around.
Built like family
CFactory shares the family skeleton — Python 3.13 + FastAPI, React 19 + Vite, PostgreSQL, a Nix dev shell — and reuses AIFactory’s enterprise auth so security and operations match from day one.
We’re building it in the open, spine-first. Follow along on GitHub and on the roadmap.