Idea → Source of Truth

The distance between a raw idea and a plan you can build.

MADHOUSE turns a spoken thought, a photo, or a scribbled note into a structured, trackable project — an orchestrated workspace where AI does the shaping and you stay the decision-maker.

6 orchestrated AI roles Self-hostable — your keys, your data Credential-blind by design
yourname.madhouse.app / source-of-truth
1 · Original Idea 2 · Roadmap 3 · Resources ◱ Canvas
Active Knowledge Grounded
Self-publishing ebooks as a 2026 side hustle
The raw idea you captured — refined by the associate into a clear objective, then expanded with roadmap, resources, and next steps.
● building Content audit & categorization
● done Deduplicate source list
● planned Draft the launch checklist
Orchestratorready
01 · Classifydone
02 · Build promptdone
03 · Roadmappending
The problem

Ideas die in the gap between having them and doing them.

Everyone captures ideas. Almost no one has a reliable way to turn a raw spark into something structured enough to act on — before the momentum is gone.

01

Capture isn't the problem

Notes apps, voice memos, and docs are full of half-formed ideas. Capturing is easy. Nothing turns that raw input into a plan you can execute.

02

AI tools give you a wall of text

General chatbots respond, but they don't hold structure. You get an eloquent answer, not a tracked, evolving source of truth you can return to and build against.

03

Every idea needs something different

One idea is mostly a prompt for a tool. Another is a sequencing problem. A third is "what do I even need." Nothing diagnoses which — it treats them all the same.

Why now

AI can finally do the shaping — not just the answering.

Models are now good enough to diagnose an idea's real shape, ask the right clarifying questions, and maintain a living structure. The workflow was impossible two years ago. It's the whole product now.

The old way

A chat window and good intentions

  • Paste an idea, get a wall of prose you'll never revisit
  • No memory of structure — every session starts over
  • You do the organizing, the AI just talks
  • No way to track what's done, building, or planned
The MADHOUSE way

An orchestrated workspace with a spine

  • An idea becomes a structured, tabbed source of truth
  • Roadmap, resources, and status persist and evolve
  • Six AI roles shape it; you approve every change
  • Progress tracked from the small pieces up, automatically
How it works

From input to source of truth, in one guided pass.

You bring the idea in any form. MADHOUSE clarifies it, shapes it, and hands you a workspace you can steer — never a black box.

STEP 01

Capture

Type it, paste it, or upload it. However the idea arrived, it goes into MADHOUSE as-is.

STEP 02

Associate clarifies

Before anything is built, a front-line AI asks up to four sharp questions — only when the idea is genuinely ambiguous. A clear idea flows straight through.

STEP 03

Orchestrator shapes

The clarified idea becomes a source of truth: original idea, roadmap, and resources — each a living document you can mark and edit.

STEP 04

You steer

Set a goal and get a ready-to-run work order, or propose a change the AI drafts and you approve. Nothing changes without your say.

The product

Six AI roles, one workspace, always under your control.

MADHOUSE isn't one chatbot. It's a small, specialized team — each role with a distinct job, a distinct temperament, and hard boundaries on what it can touch.

The orchestrator

The one that shapes the work

Reads the whole project, classifies the idea's nature, and assembles the source of truth — weighting its effort toward what the idea actually needs, whether that's prompts, a roadmap, or resources. It proposes; you dispose.

Advisor

Pink Team

Suggests the easiest, highest-leverage next step to keep momentum. Advice only — it never edits your documents.

Adversary

⚔ Red Team

Grounded, adversarial scrutiny. It surfaces the most important risk to address next — ranked by consequence, not by ease. Deliberately different from Pink Team, so you see both the safe path and the sharp one.

Diagnostician

Debugger

Reads your repositories to advise and author fixes — but has no ability to write or deploy on its own. The write path simply doesn't exist in its wiring. Assistance without self-escalation.

Reporter

Where it stands

A read-only status rollup that reads across your roadmap, marks, and repo to tell you what moved, what's building, what's blocked, and what's next. It reports — it never changes a thing.

Guide

Help

Explains MADHOUSE in plain language, on demand. Ask what a role does or how a step works and get a clear answer — so a newcomer is never stuck, and never has to leave the workspace to figure it out.

Under the hood

Not one AI wearing four hats — a real model registry.

Each role runs on its own API key and its own model, chosen for the job it does. That's an infrastructure decision most "AI apps" skip, and it's a quiet part of what makes MADHOUSE feel like a team rather than a single assistant.

RoleModel tierWhy it runs separately
ASSOCIATE Balanced The front door. Reads a raw idea and decides what to ask — tuned for judgment, not verbosity.
ORCHESTRATOR High-capability Shapes the whole source of truth and classifies the idea's nature — the heaviest reasoning, on its own budget.
FORGE ×3 Task-specialized Three document agents build prompt, roadmap, and resources in parallel — isolated keys so one stage can't starve another.
RED TEAM High-capability Adversarial scrutiny needs a strong model and a distinct temperature — kept fully separate from the advisor so their views never blur.
PINK TEAM Fast Quick, momentum-keeping suggestions. A lighter model is the right tool — and its own key keeps it cheap to run often.
DEBUGGER High-capability Reads code to advise and author fixes. Strong reasoning, its own key — and, by design, no write path of its own.
REPORTER Fast A read-only status rollup across roadmap, marks, and repo. A lighter model is right for synthesis — and it never mutates anything.
HELP Fast Answers "how does MADHOUSE work" in plain language. Quick, cheap, its own key — a guide that's always one question away.
INTAKE + RESEARCH Multi-provider Idea intake and optional web research run on separate providers — so MADHOUSE isn't captive to a single vendor.

Right model for each job

Heavy reasoning where it matters, fast and cheap where it doesn't — instead of paying premium rates for every call.

Isolated budgets

Separate keys mean per-role cost visibility and spend limits — one runaway role can't quietly drain the rest.

Contained blast radius

A revoked or rate-limited key degrades one role, not the whole product. The system keeps running.

Provider-swappable

Roles resolve their model through a registry, so upgrading a model — or a whole provider — is a config change, not a rebuild.

What it costs to run

You pay for what you use — and it's built not to overpay.

Most people reach for AI through flat monthly subscriptions, and stack several of them. MADHOUSE runs on your own API keys, metered by actual use — and because each role picks the right-sized model, it spends on capability only where the work demands it.

The usual waystacked flat subscriptions ~$110/mo, 5 standard tiers
One flagship chat subscriptione.g. a single $20/mo plan $20/mo flat, capped
Power/premium tierswhen you hit the limits $100–$300/mo flat
MADHOUSEbring-your-own-key, metered Pay-per-useprovider rates, no markup

Metered, not flat

Idea work is bursty — a flurry when inspiration hits, then quiet. Metered API usage fits that shape far better than a flat monthly fee you pay whether you use it or not.

Right-sized model per role

Premium capability costs many times more per token than a fast model. MADHOUSE sends only the heavy reasoning to top-tier models and the light work to cheaper ones — capturing that gap by design.

No subscription stacking

One workspace orchestrates several models. You're not paying for a separate seat on every tool just to have the right model available for each kind of task.

No reseller margin

Because you connect your own provider keys, you pay the provider directly. There's no platform markup layered on top of your usage.

Honest framing: bring-your-own-key means your cost tracks your usage, so a very heavy, all-day user could spend more than a flat plan — while a typical bursty user usually spends less than stacking subscriptions. The durable advantage isn't a fixed discount; it's that you only pay for what you actually run, at provider rates, with the architecture tuned to avoid paying premium prices for routine work. Subscription figures reflect standard public pricing as of mid-2026 and will change.

Why it's defensible

The moat is the discipline, not the model.

Anyone can call an API. What's hard to copy is a system where AI is genuinely powerful and genuinely safe — where every capability is fenced, every action reversible, and every secret protected. That architecture is the product.

ONE-DIRECTIONAL SAFETY

The AI can act — you're never surprised

A user action never silently triggers an AI cascade, and AI changes to your source of truth are proposed, previewed, and confirmed before anything is written. Every mutation is a single bounded pass — no runaway loops, no black-box edits.

CREDENTIAL BLINDNESS

Your secrets never reach a model

Private keys and values live in a vault that is masked, owner-scoped, and structurally excluded from anything an AI role can see. The agents are blind to your credentials by design, not by policy.

REVERSIBLE BY DEFAULT

Every AI change can be undone

The system snapshots before it acts, so any AI-driven change to your documents can be reversed with one confirmation. Powerful assistance you can always walk back.

BOUNDED CAPABILITY

Roles that can't overstep

The Debugger can read and advise but has no write path. Advisors can suggest but never edit. Capability is granted structurally — an agent can't do what it isn't wired to do, no matter what it's asked.

Who it's for

Meet it where you are — from your first AI project to your hundredth.

MADHOUSE grows with you. The same workspace that walks a newcomer through their first idea becomes a power tool the moment you're ready to run it your way.

🧭
New to AI

You have the ideas — you're still learning the tools

You don't need to know what a "prompt" is or how any of this works under the hood. MADHOUSE asks you plain questions, shows you the easiest next move, and never changes anything without your yes. You learn the shape of the work by doing it, guided the whole way.

Built for you: the Help agent that explains MADHOUSE in plain language, and Pink Team pointing you to the single easiest, highest-leverage next step — so you're never staring at a blank screen.
⚙️
Working developer

You've got projects in GitHub — and no real record of them

The code exists; the documentation, the status, the "what was I even building here" doesn't. MADHOUSE reads your repositories, tracks where each project actually stands, and keeps a living source of truth so you can pick any project back up without the cold-start tax.

Built for you: direct GitHub integration, per-project status tracking, and the Debugger reading your attached repo to advise and author fixes — without ever writing or deploying on its own.
Advanced & self-hosting

You know exactly what AI can do — now make it yours

You don't need hand-holding; you need control. Buy MADHOUSE, run it on your own keys and infrastructure, and extend the spine with your own specialized agents. The adversarial roles aren't training wheels for you — they're an edge, surfacing the sharp risks a single assistant would smooth over.

Built for you: a real agent registry you extend with custom roles on their own models and keys, plus Red Team's grounded, adversarial scrutiny — the power-user advantages a single chatbot can't offer.
Own it completely

Self-hostable, single-tenant, private by default.

MADHOUSE isn't a walled garden. A technically capable user can stand up their own fully independent instance — their own keys, their own data, gated to their own account. A guided in-app wizard validates each credential and hands off cleanly to deploy.

  • Owner-locked — only you can reach your instance
  • Runs on a single API key; providers are swappable
  • Credentials tested live, never stored or logged
  • One-click deploy, then a guided setup wizard
# deploy your own MADHOUSE
$ click "Deploy with Vercel" →
forking to your GitHub…
provisioning your instance…
instance live (unconfigured)
$ open /setup → guided wizard
your keys · your data · your rules
Own it · extend it

A base you own — and build your own agents on.

MADHOUSE is the upstream orchestration and source-of-truth layer other tools plug into — and it's yours to run. You get the codebase, host it on your own infrastructure, and extend the spine: the multi-agent registry is a real extension point, so teams add their own specialized roles instead of being locked to the built-in set.

# add your own agent to the spine
1 register the model → AGENT_MODELS.scout
2 give it its own key ANTHROPIC_KEY_SCOUT
  (falls back to the shared key)
3 write its prompt + route /api/scout
runs on its own model + budget, fenced like the rest

A base, not a black box

You get the full codebase and run the orchestration + source-of-truth layer yourself. It's a starter to operate and shape — not a service you rent a seat on.

Add your own agents

Each role is a registry entry with its own key and model. Adding one is a registry slot + a key + a prompt — so you extend MADHOUSE with your own specialized agents. A clean, modular codebase — not a plugin store or a one-click install.

Own your keys, costs & data

Credential-blind by design and bring-your-own-key throughout. Run on your own Anthropic, Supabase, and GitHub accounts — your spend, your data, and provider-swappable, with no lock to a single model vendor.

Self-host in minutes

A documented path, already shipped: a Deploy-to-Vercel button, a guided setup wizard, and an owner-lock that keeps the instance yours. Not vaporware — it's built and running.

Honest framing: "extend it" means code-level extension in a clean, modular codebase with an agent registry and per-agent keys — not a plugin marketplace, an agent store, or a one-click agent install (those don't exist, and we won't pretend they do). Self-host is a starter you run and own on your own keys — not a managed, turnkey hosted product. Licensing and modification terms are set directly with us.

The model

Two ways to run it — you choose the tradeoff.

A hosted product for people who want zero setup, and a self-host license for people who want total control. Bring-your-own-key throughout, so usage costs stay with the user.

Own it
Self-Host
BYOK / your infra
The base starter — run the full codebase on your own infrastructure, own your keys and data, and extend the spine with your own agents.
  • Full codebase, your infrastructure
  • Add your own agents — registry entry + key
  • One-click deploy + guided setup wizard
  • Bring-your-own-key — your costs, your data
Planned
Hosted Pro
Subscription / seat
We host it, you bring your keys. Zero setup, private workspace, always current.
  • Managed, always-updated instance
  • Bring-your-own-key — no usage markup
  • Priority support & onboarding
Agency / Team
Custom / contract
Per-client private instances for consultancies deploying — and extending — MADHOUSE for others.
  • Client-scoped instances
  • Deployment tooling & templates
  • White-glove setup

Self-Host is the first-class offering — you run and own the instance, a starter you operate rather than a service you rent. Pricing shown is directional — the model, not a rate card. Bring-your-own-key means the customer pays their own AI usage, keeping our margins clean. Licensing and modification terms are set directly with us — talk to us.

Where it stands

Built and working. Pre-launch, by choice.

MADHOUSE is a functioning product, not a mockup — the workspace, the six AI roles, the safety architecture, and the self-host path are all built and deployed. What's ahead is customers, not core engineering.

Live
Deployed and running — the full workspace, agents, and self-host wizard are shipped.
6 roles
Orchestrator, Pink Team, Red Team, Debugger, Reporter, and Help — each fenced, each specialized.
Day 1
Self-hostable from the first release — the go-to-market can start the moment the story does.

Honest framing: this is a pre-revenue, pre-traction stage. The pitch is the product, the architecture, and the defensibility — the numbers come next.

The opportunity

Back the workspace that turns ideas into things you can build.

The capture market is saturated. The "turn capture into structured, safe, trackable action" market is wide open — and MADHOUSE is already built to own it. Let's talk about what's next.