How it works
A copy-on-write filesystem for running many agents against one repository. The full technical description of ctxfs — what it does, what it is made of, and what state it is in.
The premise
Running N coding agents against one repository on Windows normally means N full working copies. ctxfs gives each agent a private, writable view of the same repository instead: reads fall through to a shared read-only base snapshot, and writes are captured copy-on-write into a per-context upper layer.
Each context is a real git repository, so work is reconciled back through ordinary git merges rather than a bespoke sync protocol. The base is pinned at one commit and is never written to.
What is included
Shared read-only base snapshot with per-context copy-on-write upper layers
Every context is a real git repository — reconciliation is a normal merge
Passthrough, overlay, git lifecycle and hardening milestones complete
End-to-end acceptance demonstration passing against the specification
How it is built
ctxfs is built on Rust, Windows filesystem APIs and git.
It carries 66 commits since Jul 2026, most recently changed in Jul 2026.
The method behind it
Every project here starts as a written specification, and the specification stays authoritative once code exists. The repository keeps a build plan with live status, an append-only decision log recording where the implementation departed from the plan, and a handoff document describing the true current state — including the parts that are not finished.
That is why a portfolio this wide is not a shallow one, and it is why this page can describe the product without overstating it.
Current state
Prototype. Working, not yet finished.
The source is private. What this page claims is checkable against the product itself, not the repository.
Next
Put ctxfs in your pipeline
Ask about integrating it with the repositories and agents you already run.