Turn your plan into
pull requests.

Confab is a desktop app that connects to your board and builds the stories on it. Dispatch a card and your own Claude builds it on your machine, pushes a branch to your own git host, and opens a pull request with the cost, tokens, and time reported back to the card. Nothing merges without you.

// bring your own Claude · runs on your machine

acme-api
BD
Board10:14 AM
Card #218 moved to Queued: “Return a Retry-After header with every 429.”
C
Confab10:14 AM
Picked up #218. Building on branch confab/218-retry-after, right here on your machine.
Building locally
C
Confab10:21 AM
Done. Self-reviewed against the acceptance criteria, pushed the branch to your git host, and opened a pull request. Review PR →
C
Confab10:21 AM
$0.84 · 41.2k tokens · 7m 05s, reported to the card.
Yours, end to end
macOS Windows Your own Claude Your repo, local Your git host (Forgejo, GitHub) Foothold Foundry board Jira, Linear, Shortcut (coming soon)

From plan to pull request

Confab is the execution arm of your planning board. You decide what to build, it does the building, and you review everything that ships.

The board is the interface

Plan the way you already do: cards on a board. When a card is ready, drop it in Queued. Confab picks it up with the full context behind it: the story, the acceptance criteria, and the architecture decisions your team already made.

It builds against your real codebase, in a workspace bound to your repo folder. No copy-pasting tickets into a prompt.

Execution Board · acme-api
BD
Board #218 “Return Retry-After with every 429” moved to Queued
C
Confab Picked up #218 with 4 acceptance criteria and 2 linked architecture decisions. Building.
C
Confab Branch confab/218-retry-after pushed. Pull request open. Review →

Review-sized stories, real review

Confab builds work units sized for a human to read: one card, one branch, one pull request. It self-reviews the diff against the card's acceptance criteria before handing it over, and then it stops. Nothing merges itself.

If a card is too big to review comfortably, it's too big to dispatch. Small cards are how you stay in charge of an AI that codes.

"AI-generated PRs wait 4.6x longer for review. Feature branch throughput is up 59% but actual shipping fell 7%."
- 2026 industry research. Review-sized cards are the fix.
Pull request · your git host
C
confab/218-retry-after +64 −12 across 3 files
Self-review 4 of 4 acceptance criteria met, tests added for the 429 path
YOU
Status Waiting on your review. Merging is your call.

Every build reports what it cost

When the pull request opens, the card gets the receipts: what the build cost in dollars, how many tokens it used, and how long it took. No credit meters, no mystery math.

Over time your board becomes its own cost model. You can see what a story of a given size runs, and plan with real numbers instead of vibes.

Build telemetry · acme-api
C
#218 Retry-After header · $0.84 · 41.2k tokens · 7m 05s
C
#214 Webhook retries · $1.12 · 58.9k tokens · 9m 41s
C
#209 Session cookie flags · $0.31 · 17.4k tokens · 3m 58s

Your stack, your keys, your call

Confab runs your own Claude under your subscription or API key, on your machine, and pushes to the git host you already run: self-hosted Forgejo, GitHub, and friends. Nothing about your setup has to change.

And when you want to steer, each workspace has its own chat with the agent. Course-correcting a build is a message, not a meeting.

Workspace chat · acme-api
DV
Jake On #218, reuse the middleware pattern from #204 instead of a new helper.
C
Confab Got it. Reworked on the same branch and refreshed the pull request. Review →

An execution loop you can trust

Not another chat bot. A desktop app that ships the cards on your board, on your own machine, one reviewable pull request at a time.

🌐

Your board, executed

Mark a card ready and Confab picks it up with its full context: the story, the acceptance criteria, and your architecture decisions. Your backlog becomes a build queue.

📁

Review-sized, human-reviewed

One card, one branch, one pull request, sized so a person actually reads the diff. By default nothing touches main and nothing merges itself; the PR waits for you.

🤖

Runs on your machine

A per-workspace bridge runs your own Claude locally against one repo folder. One window per project, nothing bleeds between them, and your code never leaves your computer.

🔒

Bring your own Claude

Use your existing Claude subscription or API key. Your quota, your bill, no markup, no proxy, no reselling your tokens. Compliant with provider terms by design.

📊

Telemetry on every build

Every card reports what it cost: dollars, tokens, and wall-clock time, posted back to the card when the pull request opens. You always know what your board is spending.

🛡

Your own git host

Branches and pull requests land on the git host you already run: self-hosted Forgejo, GitHub, and friends. Your code never has to move anywhere new.

Use Cases

See what teams run through Confab

Real cards, real pull requests. Not one person babysitting a terminal: a plan turning into reviewed changes.

Board: #218, #219, #221 moved to Queued
Confab: Building #218... PR open. $0.84 · 7m
Confab: Building #219... PR open. $0.52 · 5m
   3 PRs waiting for your review →
Founder+AI

Ship the small stuff while you do the big stuff

Queue three small cards before lunch, review three pull requests after. The backlog burns down while you stay on the work only you can do. Every change waits for your review.

Card #226: Login 500s after OAuth redirect
Confab: Found it · callback skips the state param.
   Fixed on confab/226-oauth-state + regression test
   PR → $0.41 · 22k tokens · 4m
Dev+AI

Bug card in, fix and regression test out

Write the bug up as a card the way you already do. Confab reproduces it against the real code, fixes it on a branch, adds a test, and opens a pull request with the cost on the card.

Sprint 14 · 11 cards executed
Total: $9.60 · 512k tokens · 1h 22m build time
Merged: 9 · sent back with comments: 2
   every change branch + PR, zero direct to main
Lead+AI

Know what every card cost, and what merged

Telemetry lands on each card: dollars, tokens, wall-clock. Your board becomes a real cost model per story size, and the branch-and-PR default means the audit trail is your git history.

Card #230: Export orders as CSV. 5 acceptance criteria.
Confab: Built. Self-review: 5 of 5 criteria met.
   streaming export, UTF-8 BOM, header row, date filter, empty-state
   PR → criteria checklist in the PR description
PM+AI

Acceptance criteria that actually get checked

The card's acceptance criteria are the spec. Confab builds against them and self-reviews the diff criterion by criterion before the pull request opens. Good cards in, checkable PRs out.

Card #233: Fill coverage gaps in checkout module
Confab: Found gaps: coupon validation, multi-currency, webhook error path
   9 tests added · coverage 67% → 84%
   PR → payments/checkout.test.ts
QA+AI

Coverage gaps become cards, cards become tests

Make test debt visible on the board instead of a wiki page. A coverage card gets built like any other story: Confab reads the source, writes the missing tests, and opens a PR you can inspect.

Remote: git.acme.internal (self-hosted Forgejo)
Confab: Branch pushed to your remote. PR opened via your token.
   code built locally · nothing uploaded to us
Lead+AI

Self-hosted git stays self-hosted

Confab pushes branches and opens pull requests on your own git host: self-hosted Forgejo, GitHub, and friends. The build runs on your machine, so your code never transits our servers.

Comparison

Confab vs. the other ways to run an AI that codes

Cloud agents run your work on someone else's machine. A terminal session needs you watching it. Confab connects your plan to your own AI and hands you pull requests.

CategoryConfabCloud coding agentsClaude Code in a terminalChat coding bots
Where the work runsYour machineVendor cloud, code uploadedYour machineVendor cloud
Driven by your planBoard cards in, PRs outOne prompt or ticket at a timeYou type each taskOne chat thread at a time
Work unitReview-sized story cardOften whole-feature, hard to reviewWhatever you ask forWhatever you ask for
Review gateBranch + PR by default, nothing merges itselfVaries by vendor and configYou are the gate, liveVaries
Works while you do something elseYes, until the reviewNo, it's interactivePer message
Who owns the AI billYour subscription or keyConfab runs your own Claude under your subscription or API key. Your provider bills you directly at standard rates. Confab never marks up your AI costs.Credits, usually with markupYour subscriptionVaries, often credits
Cost visibilityDollars, tokens, time on every cardA credit meterPer session, if you look
Your own git hostForgejo, GitHub, and friendsOften GitHub-onlyVaries

What to know about AI agent pricing

$

Credit systems resell your tokens

Cloud agents typically buy model capacity wholesale and sell it back as opaque credits. Real-world reports show $200-400/month in sustained use, even when the headline price says $50. Confab runs your own Claude, so there is nothing to resell.

$49/mo for five users on Confab Pro
$

Per-seat and per-run pricing multiplies

Some agent platforms charge per seat, per parallel run, or both. Five people running builds means five meters. Confab charges a flat team price, and the AI bill stays whatever your provider already charges you.

Flat team pricing, your own AI quota

AI provider policies are shifting

Providers keep changing how they price and enforce third-party tool access, and some have suspended accounts without warning. Confab runs your Claude under your own login, the pattern providers actually sanction.

Your subscription or API key, compliant by design
$

Most agents hide the meter

An agent that loops overnight on real money is a budgeting problem you find out about later. Confab reports cost, tokens, and wall-clock time back to every card, so the spend is visible per story, as it happens.

Telemetry on every build

Confab keeps it simple. Bring your own Claude subscription or API key and pay Anthropic directly at standard rates. We charge a flat team price, and every card shows what it cost to build. No markup, no credits, no surprises.

The math speaks

Solo? $9.99/mo. Team of 5? $49/mo.

The same team on typical per-seat pricing: $75–400/mo

Pay annually and save 20%. You bring the AI. We don't mark up your provider costs.

From card to pull request.

No servers. No terminal. No IT ticket.

1

Install the desktop app

Download Confab for Mac or Windows and open it. Everything runs locally; there is nothing to deploy.

2

Open a workspace

Point a workspace at a repo folder on your machine and connect your board. Your own Claude runs inside it, against your code.

3

Dispatch a card

Drop a card into Queued. Confab builds it on a branch, self-reviews it against the acceptance criteria, opens a pull request, and posts cost, tokens, and time back to the card.

Security isn't a feature. It's the foundation.

Confab runs on your own machine and touches your code. We take that seriously. Every claim here is backed by an architectural decision, not a marketing checkbox.

Transparent pricing. No surprises.

Start free. Bring your own AI. We don't mark up your provider costs.

Monthly
Annual SAVE 20%

Free

$0

See what Confab can do

  • 1 user
  • 1 project, 1 channel
  • Slack or Discord or Telegram + Web
  • Ask questions, edit documents
  • 7-day session history
  • /compare, code editing
Join Waitlist

Personal

$9.99/mo

Solo power user

  • 1 user
  • 3 projects, unlimited channels
  • All channels including SMS
  • All permissions (code, docs, workflows)
  • /compare (2 models)
  • Auto-repo creation
  • 30-day session history
Join Waitlist

Enterprise

$199/mo

20 users included

  • Everything in Pro
  • 20 users (volume discounts available)
  • Unlimited projects
  • Unlimited session history
  • Per-user & org-wide budget controls
  • Priority support
Join Waitlist

Stop planning cards. Start shipping them.

Confab turns the cards on your board into reviewed pull requests, built on your own machine with your own Claude, with the cost reported back to every card. You stay in control: every change is a branch, every branch is a pull request, and nothing lands until you say so. That's Confab.

Get Started FreeSee Pricing

// bring your own Claude · runs locally · your code stays yours