For developers who use AI coding agents

Always know what to
focus on next.

We help developers solve scattered work across different tools by integrating them into one place, so you have full visibility across multiple projects and can work seamlessly with coding agents.

Early access · Built with working developers who use AI coding tools every day
fables in Project mode: the project home with GitHub, CI signals and activityfables in Split mode: the project on the left, its terminal tabs docked rightfables in Terminal mode: the terminal full screen
Works with the tools you already work with
GitHubLinearJiraSupabaseVercelSentryPostHognpmExpoClaude CodeACP
One place, full visibility

Everything that needs you, in one prioritised view.

Fables brings your tickets, reviews, errors, logs, deploys and agent runs from the tools you already use into one place. You always know what to focus on next, and you can hand the right context to your agent in one click. Then it shows you what all that work says about how you are growing as a developer.

Always know what to focus on next

Tickets, review requests, errors, failing builds and agent runs from the tools you already use land in one prioritised list. Each item says why it is there, so you can see what needs you now and what can wait.

01Prioritised for you: what blocks others, what is due today, what broke after your deploy.
02Every item carries a short reason, so the order never feels random.
03New signals, like an error report or a review request, appear as soon as something needs you.
Sentry
Checkout crash: null dispatch
142 events · 12 users in the last hour
PostHog
Signup drop-off on step 2
Conversion down 18% week-over-week
Vercel
Preview build failing
Type error in app/api/route.ts

Hand it to your agent, context included

Pick an item and Fables packs the context your agent needs: the ticket, the diff, the error trace and your project notes. No more gathering it by hand from four applications before every task.

01One click starts your coding agent on its own branch, with the context attached.
02Start from a plan you approve first, or go straight to the fix.
03You review the diff and merge with gh. Nothing proprietary in the loop.
TKT-204 Refund flow via Stripe PR ready
TKT-205 Cache availability at the edge agent running
TKT-206 Waitlist when a slot frees open
SIG-12 Fix checkout crash agent running

Several agents, one review queue

Run agents in parallel, each in its own git worktree, so nothing collides. Finished runs come back into the same list, ready to review: the diff, the tests and what changed.

01Each agent gets its own worktree and branch.
02Send a run back with a note and the agent picks up where it left off.
03Agent runs sit in the same prioritised list as everything else, not in another window.
#152 Refund flow via Stripe 58 tests ✓
#151 Cache availability at the edge ready
#150 Waitlist when a slot frees sent back · 1 note
TKT-207 Partial refunds: e2e cover running 4m
4 dispatched overnight · 3 waiting on you

A terminal that's just a terminal

Edge-to-edge, full-screen when you want it. Real git, gh, npm and claude. Nothing to relearn, nothing flaky to wrap. The agents drive the same shell you do.

01No proprietary CLI. What you type is what ships.
02SSH straight into GitHub Codespaces when you sign in with GitHub.
tidepool · ~/dev/tidepool · main
$ git worktree add ../tidepool@TKT-204 -b feat/refunds
▸ ../tidepool@TKT-204 · branch feat/refunds
$ claude --agent --task TKT-204
agent · acceptance criteria ✓ · writing tests
✓ 58 passed, 0 failed
$ gh pr create --fill --base main
✓ PR opened · ticket → review

Bring any agent. fables speaks ACP.

Fables works with the coding agent you already use, so it adds to it rather than replacing it. Claude Code is native today: sessions, skills, tools and subagents are first class. The rest arrives over the Agent Client Protocol, the open standard between editors and agents, so adding an agent becomes a connection rather than a rewrite.

01Claude ACP is the default and the only agent enabled at launch.
02Codex, Gemini CLI and your own agent land next, in the same list and the same review.
03Because it’s an open protocol, a run can move between agents without losing context.
claude code ACP · native · skills, subagents default
codex over ACP coming soon
gemini cli over ACP coming soon
your agent ACP + SDK coming soon

Get better at your craft, from your own work

Fables looks at the work you actually do: the code you write, the reviews you give, the agents you direct. It shows where you are strong and what is worth learning next. Skill growth for the developer, not a report card for the manager.

01Built from real work: your commits, reviews and agent runs, not quizzes or self-assessments.
02Names your developer archetype, a short profile of how you work, and suggests the next thing to learn in the context of your own codebase and stack.
03Yours first. You decide what goes on your developer profile and what stays private.
strong API design · test coverage · agent briefs from 31 PRs
growing Reviewing agent output for edge cases 3 misses caught late
next Stripe webhooks: idempotency comes up in TKT-204
Private to you · share to your profile when you want
Your tools, connected

Everything around the code, in one place.

Σ

Skills from the store

Add official Claude skills (Next.js, Supabase RLS, Sentry triage) and your agents follow your conventions automatically.

≡

Your tickets, ready to run

A board with room to read: description, acceptance criteria and estimate on the card. Tick several, dispatch them in parallel, review them as one queue.

∑

Worktrees, managed

See every active branch and agent at a glance. Open, review, or clean up worktrees without leaving fables.

⌃

SDK & CLI

Script the workspace. Dispatch runs, read the record, and wire fables into CI with a typed SDK and a fables command.

ƒ

Your stack, connected

GitHub, Linear, Jira, Supabase, Vercel, Sentry, PostHog: connected once, used everywhere in Fables.

•

Product, defined once

Specs, goals and screenshots live in Fables and sync to your repo, so every agent works from the same brief.

“Get more done, get better as you go.”
Code, tools and agents, together in one place
Free during early access

Bring your tools into one place.

Register your interest and we’ll email you the moment your platform is ready.

Windows & Linux builds coming soon.

Launches soon
macOS
Apple silicon & Intel · macOS 13.0+