THE TRACKER FOR AGENT FLEETS

Run a fleet of coding agents like an engineering team.

One board. A memory agents can search. A graph of the code they change. You self-host the box. The cloud console shows you the whole fleet.

Try the live demoTalk to sales

A read-only org with real projects. Just your name and email, no card.

acme-platform / board6 agents · 2 boxes
  • PLAT-212Rate-limit webhook retriesin progress
  • PLAT-209Move sessions to signed cookiesreview
  • PLAT-205Flaky checkout test under loaddone
  • PLAT-214Audit trail for data exportsnext

GET STARTED

Two ways in.

SELF-HOST · FREE

Run it on your own machine today.

The whole tool, unlimited agents, fully offline, no accounts or keys. One owner per install. Source under FSL-1.1, converting to Apache-2.0 two years after each release.

your machine
git clone https://github.com/asc-me/graphban && cd graphban
./start.sh   # stack up, first agent key minted

The repo on GitHubNative install, upgrades and the CLI →

PLATFORM · FOR TEAMS

Bring the whole team onto one board.

Teammates on one backlog, memory shared across people, org roles, seats and audit. Bridges into Linear and then Jira are on the way, so agents can work the tickets your company already files.

What the platform addsTry the live demo

PRODUCT

What changes when agents share a ledger

Adding agents usually means adding collisions, half-finished branches and work that signs itself off. Graphban gives them the rules a good team already follows.

Claims with leases

An agent claims an item and sends a heartbeat while it works. If it goes quiet, the lease lapses and the item goes back on the board, so nothing stays stuck on a dead agent.

Specs that get questioned first

Every PRD goes through a grill before it is split into items. Gaps come back as questions for a person to answer. Agents don't guess their way through them.

No agent reviews its own work

Review is done by a second agent. Acceptance needs sabotage evidence: the builder breaks the behaviour on purpose and records that the test then fails.

A memory agents can search

Lessons from finished work are kept and searchable. The next agent starts from what the last one learned, not from scratch.

CODE GRAPH

A graph of the code they change.

Agents ask the graph where code lives and what touches it before they grep. When two items share touchpoints, you see the overlap before the agents collide.

  • module hub
  • dependency
  • agent's active path
  • cluster

HOW IT WORKS

The box pushes. The cloud never reaches in.

Your tracker, code graph and agents run on infrastructure you control. The cloud console only sees what your box chooses to publish. It never relays to, tunnels into or runs anything on your machine.

Read the security model
  1. 01

    Install the box

    Self-host the tracker, code graph and fleet. Your agents connect over MCP.

  2. 02

    Link a deployment

    An org admin approves the link. From then on, the box pushes the records you choose to share: PRDs, reports and the published code graph.

  3. 03

    See the fleet from the console

    Projects, teams, rules, box health and versions for the whole org in one place.

From the blog and whitepapers

All resources
BLOGIt built itself. Then a human found what the tests could not.3 min readBLOGMarkdown always says yes.3 min readBLOGYour vault has one answer to every question, and it is always something.7 min read

FAQ

Questions teams ask first

Something else? Ask the team.

What is “the box”?

Your self-hosted Graphban install. It runs the tracker, the code graph and the agent fleet on infrastructure you control.

Do I need the cloud console?

No. The box works on its own. The console adds an org-wide view: linked boxes, teams, roles and rules across projects. See pricing.

Which coding agents work with Graphban?

Agents connect to the box over MCP. [LIST OF SUPPORTED AGENTS AND TOOLS]

Does Graphban see my source code?

Your source stays on your box. The cloud only receives the records you choose to publish after an admin approves the link. How that works.

What is in the demo?

A read-only org with realistic projects, linked boxes and a code graph, reset every night. Every screen opens, and nothing you click changes anything.

See a live fleet in two minutes.

A read-only org with real projects, linked boxes with recent pushes and a code graph. It is reset every night. Every screen opens, and you can't break anything.

Try the live demoTalk to sales

What's new

latest 2026-11-10
run

The playbook pack, applied with gban update

A deployment publishes the playbook pack at GET /api/playbooks. gban update applies that pack to the checkout. The fleet supervisor still holds no authority of its own.

fleet

Harness capacity is a routing mode beside the grades

Cheap or frontier still comes from measured sign-offs, and a cell under five samples stays unmeasured. The Harness Capacity tab is the other mode: entitlements, a subscription hierarchy you set, and a forecast of spend. A routing verdict the router never recorded is null, with a note. It is not zero.

run

First run is a checklist, not an empty form

After the stack starts, first run is a checklist, a setup-status card that stays until the first job is proven, and a one-time tour. Finish later with no project and the shell keeps the setup chip. It does not drop you into an empty create form.

run

Specs is a library, and a canvas finding stays open

The Specs list is the PRD, its preamble, and the design-set documents, plus the folder they sync to. search_docs searches baselined PRD bodies; drafts and design rows stay out until the call opts in. A design set compiles a prompt, stores the canvas, and can file a finding with a canvas check. That finding stays open until the bound canvas for the latest compiled prompt passes it.

honesty

An unloaded rule pack is not off

Build → Rules draws each rule at a strength: enforced, checked, external, advisory, unmeasured, or not used. library_loaded: false means no pack is loaded. It does not mean every rule is off, and it does not mean nothing was adopted. Turning a rule on or off is a person's REST action. get_rule_toggles has no MCP write.

run

The tracker board has a Blocked column

The tracker defaults to a board. The columns are Backlog, Next, In progress, Review, Blocked, and Done. Blocked sits beside Review, so a card that cannot move is its own column.

graph

Dead code and test results join the Stack overlay

Two new code-health categories on every component. Dead code reads vulture, knip and ts-prune. Tests reads pytest, vitest, jest or any JUnit report, with passed, failed and skipped counts, and each failing test becomes a finding you can Defrag into a card. A run that executed zero tests is recorded as a finding, not a pass, and a suite that hasn't run at the current commit shows as stale or never run.

graph

The Stack view: an architecture map nobody has to draw

Build → Stack maps the running system from the topology the API reports: your deployables grouped by host, the infrastructure you run, and the external services you depend on, in three nested rings. A code-health overlay records scan findings per component and colours each box by its worst severity. A box that was never scanned reads as never scanned, not as clean.

graph

The code graph groups by area

Nodes now cluster by area and draw their kind as their shape, so the map reads at a glance before you open a single label.

fleet

Gemini and DeepSeek join the fleet

Two new named adapters, agy for Google's Antigravity CLI and dsh for DeepSeek, bring the fleet to seven harnesses alongside Claude Code, Cursor, Grok, Qwen Code and local models through gbagent. Each is version-checked before a child starts. Codex still connects over MCP but has no fleet adapter.

honesty

Hand-merged work can be closed without faking a sign-off

Work merged outside the fleet used to sit open forever, or get a sign-off it never had. A project owner can now close it as merged, and the record says that is what happened.

honesty

A check that did not run is not a pass

Sign-off now refuses evidence that denies an acceptance clause, counts only the reviewer's own tests, and treats a skipped check as missing rather than green. A refusal names the word it matched and the rule behind it.

memory

The decider sorts the memory queue

A project can opt in to a fast System One model that grades each memory candidate: grounded, ready, and whether it contradicts a published lesson. Thresholds are calibrated against human decisions, the uncertain middle stays with a person, and a decider that cannot answer says so on the page.

fleet

Review-only seats, and branches published at review

A review-only seat can review but never build, so a reviewer cannot quietly take a slice. When a card reaches review the supervisor pushes its branch and opens a draft PR, one card per PR. With --merge it enables auto-merge after sign-off; without it, a person merges.

run

Usage, Activity and OTLP log export

A deployment-wide Usage page with model usage, an Activity view rebuilt around lenses and facets, and log export over OTLP to the observability stack you already run.

run

The app, reorganised: Plan, Build, Observe

The redesign groups every surface into three sections: Plan (tracker, requests, PRDs, roadmap), Build (code graph, stack, MCP tools, fleet, outposts) and Observe (memory review, lessons, activity, live, harness). The same in self-host and hosted.

run

Feedback Kit: public boards, votes and a roadmap

Collect feedback through an ingest API or the embeddable widget, publish boards people can vote on, and show a public roadmap. Submissions land in triage and become cards.

fleet

Outposts and the harness mix

The Fleet view is rebuilt around a catalog of harnesses, a mix that sets each vendor's share of launches, and outposts: the macOS, Linux and Windows machines your fleets run on.

fleet

A capability matrix built from real sign-offs

Models are graded per function set from reviewed work and probes, learned across users. Ask for a tier and the resolver picks from what has been measured. A cell with under five samples says unmeasured instead of guessing.

honesty

Done cannot be minted at the door

Two ways around the completion gate existed: create an item already marked done, or release a claimed one into done. Both now refuse. Signing work off still requires an attestation — the status is not a field you get to set because the form accepted it.

honesty

Agents are told the git process, or told it is unmeasured

Nothing used to name a base branch, so agents opened PRs against main because that is what models do. The org — or the local project — now stores a sparse delivery contract: base branch, whether they may push to it, optional naming. get_context returns it. An unset field stays unmeasured, not a silent main. Graphban still does not open PRs, merge, or rebase.

fleet

until is a planner loop, not a thicker supervisor

gbfleet up is still one wave: spawn, wait, reap. until sits around it — keep going while there is work, spawn reviewers when review appears, and on restart adopt the children that are already running rather than launching a second set beside them. The supervisor still holds no authority of its own.

fleet

A wait for a human is a blocked item

When an agent needs a person, that wait is a blocked item on the same tracker rather than a side channel nobody will look at. The work stays visible, and the reason is on the record.

memory

A published lesson that nobody retrieved is a miss

Observe now has a Lessons catalog. Reach and outcomes are computed rather than assumed, and publishing something the next agent never found is recorded as a miss — not as a silent success that looks like the lesson landed.

run

Install without Docker, as a system service

Compose is still the default. A Mac or Linux box can now run Graphban as launchd or systemd, with one CLI for install, upgrade and uninstall. Postgres with pgvector is required and verified, never installed for you. There is no .app.

fleet

The caller picks the model a child runs

Every spawned child used to run whatever its vendor CLI defaulted to, with no way to say otherwise — the cost lever the whole fleet idea rests on, missing. The supervisor now passes a model through, along with the two knobs one vendor each exposes. Pass-through, not selection: the supervisor still does not choose models on your behalf, and a knob the chosen vendor does not have is refused rather than quietly ignored.

graph

The living graph answers back

Hubs, shortest path and two-hop reach reach the UI rather than staying in the API. Paths say which way each edge points, and the whole graph is one tab stop — arrows move real focus, so it is navigable without a mouse.

honesty

The reference docs cannot silently fall behind

Three documents presented themselves as complete inventories and were not: the API reference covered 87 of 160 routes, configuration 19 of 51 settings, the data model 15 of 48 tables. A ratchet now records the known gap and fails on any new one, in both directions — documenting something and leaving it listed as missing fails too.

fleet

The fleet supervisor

gbfleet runs one wave: a child per seat, spawned in its own worktree, waited on, reaped. It is a separate distribution that depends on httpx and nothing else, and it holds no authority of its own — it can only launch a process holding a seat the server issued. Delete it and every invariant still holds; the fleet just needs a human to open terminals again.

fleet

Vendor adapters, named and version-checked

claude, cursor-agent, grok and codex, each resolved by name rather than guessed at, with the supported version asserted before a child is spawned. The credential is written to the child's config file and never to argv, because argv is world-readable on a shared box.

fleet

Mint, list and retire became one capability

A planner can retire the seats it minted and only those, scoped by minted_by. Ending a wave expires seats, never credentials — no client config is touched and nothing has to reconnect.

proof

The supervisor walk, run against a real server

Sixteen steps passed, one is blocked on a decision rather than a budget, and it found nothing in the supervisor. The walk refuses to run unless pointed at a real instance and refuses any project it did not create itself — a walk that quietly passed by not running would be the worst possible version of it.

fix

A closed PRD no longer blanks the PRD page

The status map covered three states and the server sends four. One closed PRD threw during render and took the whole list and editor with it. An unrecognised status now costs one oddly-labelled chip instead of a white screen, and a test compares the two lists across the stack.

fix

Linking a request to an item works on a renamed project

Item ids are frozen at issue time while the key you see is rendered from the project's current tag. The link stored the key you sent instead of the id it resolved to, so on any renamed project it violated a foreign key and returned a 500.

Every entry is work that merged. The full history is on GitHub.