Currently in private beta
The harness already composed for Rails.
Turn a feature into a shipped pull request — on a new Rails app or one you already have — with agents that respect the craft and a full view of the work, the cost, and the behavior.
-
SYSTEM 20:01:48Run started · Almighty monolith · claude-sonnet-4-6
-
TOOL CALL 20:01:50Read app/models/invoice.rb
-
TOOL RESULT 20:01:51ok · enum :status draft/sent/paid
-
TOOL CALL 20:01:54Edit app/models/invoice.rb
-
TOOL RESULT 20:01:55ok · 5 lines added
-
TOOL CALL 20:01:58Bash "bin/rails test test/models/invoice_test.rb"
-
TOOL RESULT 20:02:05ok · 14 runs, 28 assertions, 0 failures
-
SYSTEM 20:02:07Opened pull request · feat/invoice-status-enum
Spec and execute — on a new Rails app or your existing one.
Foundation makes Fragua fit your codebase first: it scaffolds a new app with the maquina generators and components, or learns the conventions of one you already have. Then Spec and Execute do the work — that's the core for every project.
Optional, new app · two greenfield-only steps can run first and feed every later phase — Research studies similar apps into a synthesis you approve, then Plan turns it into an MVP plan with brand and technical guides.
Before any feature is built, Fragua gets working against your codebase — a real git repo and working tree it can build on. It's the gate: nothing downstream runs until it's ready.
Scaffolds and configures it with the maquina generators and components, plus the stack defaults a Fragua-built app ships with.
Discovers how it's already set up — your defaults and the way you work with code — and adapts to your conventions instead of imposing its own.
Describe a feature; a guided interview shapes it into a reviewable spec, then decomposes it into a task list ready to build.
Agents implement the tasks on a feature branch and open a pull request — autonomously, with the work streaming live.
This is where a spec starts.
Describe the feature in plain language — what it does and why. Attach a mock or a reference, and point Fragua at the knowledge-base files it must read. That's the whole form.
From there, a guided interview asks the follow-ups, shapes it into a reviewable spec, and decomposes it into a task list ready to build. Nothing here is a black box — you see the spec before a single line is written.
A short, specific name. Up to 120 characters.
See the attached mock as a reference.
Plain language. Up to 10,000 characters — Fragua will ask follow-ups for the rest. 128 / 10000
Context
One-off files Fragua must read — they live with this spec only, not your knowledge base. PDF, Markdown, HTML, or images.
Drop files here or click to browse
Pick the knowledge-base files Fragua must read for this spec. It reads the brief, plans, and codebase on its own.
Every artifact persists. Every later phase has access to every earlier one.
Generic agents don't know your codebase. Fragua does.
Fragua's agents ship pre-loaded with the Rails 8.1 vocabulary. Hotwire for the front end. Solid Queue for async work. Rich domain models with concerns. The One Person Framework philosophy from Rails as the default — building like one developer can ship the whole thing.
When the agent reads your code, it knows what a Stimulus controller is. When it writes a test, it writes Minitest with fixtures. When it opens a pull request, it follows the conventions your team already lives by.
- Built-in awareness of maquina_components for views
- Your Knowledge Base staged into the agent's working directory at run time
- Per-feature git worktrees for parallel work
- Pull requests opened through your host's own GitHub access
Tasks
- Data migration: drop custom amenity rows
- PropertyAmenity: require amenity
- Amenity: remove unused match
- Property model: key-based sync + publish rule
- Controller: permit amenity_keys
- Helper: amenity icons + resolver
- View: replace combobox with toggle grid
- Delete combobox Stimulus controller
- i18n es-MX: reword hint, drop placeholder
- Seeds: development.rb by key
- Fixtures: catalog-only
- Tests: model, controller, migration
- Gate + review
- Branch
- fragua/implement-amenities-selection-for-create-edit-properties-6687
- Pull request
See the work, the cost, and the behavior.
Live run timelines, token roll-ups per workspace and month, and a durable audit trail of how every agent behaved.
Begins with one developer. Indispensable for a team.
Admin and member roles, shared workspaces, and per-feature git worktrees so many agents run in parallel without stepping on each other.
Your host. Your keys. Your repo.
The control plane orchestrates; the agents run on your host against your keys. Fragua never proxies the work or sees your bill.
Orchestration, the live timeline, and cost roll-ups — the part you watch.
Your gh and Claude credentials, your repo, isolated git worktrees. Your keys and code never leave it.
Tokens billed straight to your account. Fragua can't see the bill.
AI traffic runs host → provider directly · Fragua is never in the path
Rails 8.1. SQLite. Solid Queue. Hotwire.
Fragua is built on the Maquina stack — vanilla Rails with the One Person Framework philosophy. The same conventions you'd want in a project Fragua builds for you.
The framework, modern and complete.
Five embedded databases, one server.
Database-backed async jobs.
Turbo + Stimulus, server-driven UI.
Passwordless sign-in via email codes.
Pushes and PRs through your host's access.
Flat per-account. BYOK. No surprises.
You pay Fragua a flat monthly fee for the platform. You pay your AI provider for tokens. We never sit between the two.
For independent Rails developers and personal projects.
- 1 user · unlimited workspaces
- BYOK — your AI spend, billed to you
- Full cost dashboard
- Per-feature git worktrees
For small consultancies and product teams.
- Up to 10 users · unlimited workspaces
- BYOK — your AI spend, billed to you
- Admin & member roles, shared workspaces
- Audit trail across every agent run
- Priority access to new features
Currently free during private beta · Pricing begins 30 days after public launch with email notice
Tell us what you're building.
Fragua is in private beta with a small group of teams and solo builders. We're not opening seats on a schedule — join the waitlist and we'll reach out when there's room for the work you're describing.