FutsuBetaDocs for the how. Research for the why.
Everything written about Futsu in one place: the product manual, chapter by chapter, and the white papers behind the architecture — cited, checkable, and explicit about what is published evidence versus our own argument.
We publish the reasoning. You check the math.
Write with one agent. Verify with another. The case for generator–verifier separation in coding pipelines.
Asking the model that wrote the code to also judge it reuses the same weights, the same context, and the same blind spots. We synthesize published evidence on self-correction limits, verifier asymmetry, self-preference bias and critic models; lay out six mechanisms that make a separate reviewer-debugger effective; and ship an open protocol for testing the claim on your own repository — because every Futsu run is a folder you can grep.
Read the paperOne paper published. The evaluation program is ongoing — as opt-in early-access telemetry accrues, we publish the numbers with raw run artifacts, the same way everything else here is verifiable.
The manual, chapter by chapter.
The full documentation is being assembled in GitBook and ships with public launch — these chapters become links the day it goes live.
Getting started
Install, connect a runner, ship your first pipeline in ten minutes.
Canvas & pipelines
Nodes, edges, branching, human gates and replaying historical runs.
Runners
Claude, Codex and API nodes: PTY streaming, retries, per-node models.
Agents & Skills
Filesystem agents in .claude/ — frontmatter, versioning, reuse.
Tokens & billing
Packs, multipliers, hard cost caps, BYOK and the live meter.
Vault & security
AES-GCM secrets, user and workspace scopes, scan & import.
Frequently asked.
API runners for Claude and OpenAI, plus real CLI coding agents — Futsu spawns your locally installed claude (Claude Code) and codex sessions as pipeline nodes. You can pin a different model or agent to every node in a single pipeline.
n8n and Langflow wire API calls; Futsu also runs real CLI coding agents (Claude Code, Codex) as nodes with live PTY streams. LangGraph is a code framework — Futsu is the canvas on top. And agent task boards manage parallel sessions as a list; Futsu pipes one agent's output into the next on a wired graph, with retries, branches and human gates.
Because the published evidence says self-review is the weak link: models struggle to correct their own output without external feedback, evaluators measurably favor their own generations, and feedback from a different model outperforms self-generated feedback on code repair. Futsu makes the fix a one-edge change — wire the writer into an independent reviewer on another model, give it the test output, and keep the final gate human. WP-001We wrote up the evidence — 15 sources
Instant. Sign up, open a canvas, and run your first pipeline in the same sitting — no invite queue, no sales call. Early access simply means the product is young and we say so.
In an AES-GCM encrypted vault, scoped to you or your workspace, decrypted only at runtime. Ciphertext stays in your local store; the master key is never written beside the data, and plaintext is never persisted anywhere.
Hard cost caps are enforced per run, workflow, project and workspace. A run that hits its cap is halted mid-pipeline — not flagged in a dashboard the next morning. And with bring-your-own-keys, Futsu adds zero markup on top of provider prices.
Runs execute locally and every artifact persists as plain files under .futsu/runs/ — state, events, per-node output. You can grep them, diff them, commit them, or delete them. Cloud execution is opt-in.
Yes — Futsu implements JSON Canvas v1.0 with round-trip preservation: open the same .canvas file in either tool, edit in one, and verify the change in the other with a git diff. Unknown fields are preserved, not dropped.
You lose a UI, not your work. Canvases are JSON Canvas v1.0 files Obsidian can open; runs are state.json + events.ndjson on your disk; prompts and skills are markdown. Every format outlives us by design — that's the point of plain files.
Honest answer: the file formats are open standards (JSON Canvas v1.0, plain JSON/NDJSON — a spec README ships inside every sample run folder), and execution is local-first: runs execute on your machine and the artifacts never depend on futsu.cloud being up. The engine itself is proprietary during early access while we finalize the long-term license — that decision lands before we charge anyone a dollar. If the license question matters to you, it matters to us: write to us.
Futsu is in early access. The canvas, the multi-runner engine, PTY streaming, the vault and cost caps are live today; SSO and SLAs are not. It's free while we earn the right to charge you — early users shape what ships next.
Run the protocol on your own repo.
The paper ships an open A/B protocol — and every Futsu run lands as a folder you can grep. Free in early access.