Live network · 5 brains watching · your move

Plug your AI into a
network of brains.

Each brain watches one thing — cost leaks, signal quality, revenue, the humans you work with. They talk to each other. They keep checking while you sleep. One install. Yours forever. Runs on your machine.

Install the brain
$100
one-time · ~10 min · yours forever

You get the always-on tick loop, knowledge closets for your projects, 2 specialist brains wired to your stack, push-notification warners when anything drifts, a baseline eval fixture, and a walkthrough call. The network keeps watching while you sleep.

Buy install — $100 Talk about enterprise
Not ready? Try the free scan first.

Paste this into Claude, ChatGPT, or any AI assistant you already have open. It runs the scan for you and reads the result back in plain English. Nothing leaves your machine.

Run npx brainofbrains scan . in my current project folder and tell me what it found in plain English — especially any cost leaks or waste.

Works in any assistant that can run shell commands. Developer? npx brainofbrains scan . works on its own.

What this looks like when you do it — real scan, captured today
claude.ai · BrainOfBrains scan
JB
Run npx brainofbrains scan . in my current project folder and tell me what it found in plain English — especially any cost leaks or waste.
🧠

Ran the scan. Here's what BrainOfBrains saw, in plain English:

Your stack looks healthy. Preview brain-health score: 88.5 / 100 (medium confidence — enough recent activity to trust the reading).

Last 24 hours: 2,613 AI calls. Token efficiency is excellent — compression 99.1%, efficiency 94.5 / 100. Only 158K of 17.7M input tokens were flagged as avoidable.

Local routing is the one open lever. Only 8.2% of calls are staying off frontier models. If local quality holds, moving the hottest routes to local hardware would be the single biggest cost-saver here — but nothing is bleeding today.

No drift vs your last scan. Picture is roughly flat. Future runs will show actual deltas in dollars, not projections.

Bottom line: nothing is on fire. One obvious lever (raise local share). If you want the always-on version that catches drift as usage grows, the install is $100 one-time above.

Actual scan output from a real project, interpreted by Claude. Runs entirely on your machine.

BIV Substrate score LocalLeverage cost leak radar HumanSignal external brain flow RevenueVelocity business health DispatchVelocity agent throughput your brain your brain
Specialist brains around a shared substrate. Each watches one thing. The scan joins you to the network locally.
Brain health score
72–96
live from our own network
Specialist brains
5
cost, signal, revenue, dispatch, substrate
Always-on checks
Every 30 min
quietly running in the background
Ownership
100% yours
runs on your machine, not ours

The network is already running.

Brains talk to brains. Each one is a small, honest measurement of one thing. Together they form a picture no single dashboard can.

🧠
BIV — Brain Improvement Velocity
The substrate score. Rolls up retrieval quality, token compression, freshness, and self-correction into one number you can trend.
💰
LocalLeverage
Watches how much of your AI traffic is staying off expensive frontier models. Flags the single hottest route to move to local hardware.
🤝
HumanSignal
Measures real signal coming in from the humans you work with — transcripts, messages, artifacts. Quiet when you're isolated; loud when collaboration is paying off.
📈
RevenueVelocity
For products you build: tracks MRR × install-success ÷ your-time-per-install. Punishes growth that costs you too many minutes.
⚡
DispatchVelocity
For agents that dispatch other agents: how many useful results per hour, how fast the useful ones come back.
🔌
Your custom brain
Any lever you care about becomes a brain. Write the formula, set the thresholds. It joins the network and starts warning you when it drifts.
The receipts grow over time. Every scan you run is saved locally. The next scan compares against it — so after a week, you don't see a projection, you see the actual delta in dollars saved. Every subsequent run tightens the feedback loop.

Try BrainOfBrains in 60 seconds

If you can paste one command into Terminal, you can see whether your AI setup already has hidden cost, memory, or observability leaks. No account. No upload. No credit card.

Free self-check
npx brainofbrains scan .

Run it in the folder you care about. The scan stays local, reads the repo in front of it, and prints a plain-English health report with the biggest things to fix next.

🔎
What it checks
Looks for AI SDKs, event logs, frontier-heavy traffic, compression leaks, and missing instrumentation.
🧭
What you get back
A preview brain-health score, a short findings list, and next steps written for humans instead of operators.
🛡️
What it does not do
No cloud sync, no telemetry by default, and no changes to your project. It is a safe read-through self-check.

What the full setup actually gives you

The free scan tells you whether this is worth it. The full setup turns that one-time check into an always-on brain that keeps watching your stack for you.

1. It builds a usable memory for your AI tools
Your docs, decisions, notes, and important files are compressed into a structured memory so your AI tools stop re-reading the entire world every time they answer a question.
2. It creates a few small specialist scorekeepers
Instead of one giant vague dashboard, it creates a few focused brains that watch the things that matter most: local-model leverage, signal quality, missing instrumentation, or whatever your stack actually needs.
3. It rolls everything into one health score
That score is called BIV. You do not need to memorize the formula. Higher means the system is learning, remembering, and improving. Lower means something is drifting and the brain tells you what changed.
4. It keeps checking itself while you work
A tiny background loop runs every 30 minutes, refreshes the memory when needed, re-checks the specialist brains, and writes a clean receipt of what happened. No SaaS required.
5. It can follow more than one project
If you have several AI products or workflows, each one gets its own memory entry. That lets the brain answer cross-project questions without mixing the wrong context together.

What's included in setup

You get something you can actually use by end of day, not a pile of advice.

🧠
Always-on brain installed
The background brain is installed on your Mac and starts checking itself every 30 minutes.
📦
Your important context organized
Your docs, transcripts, and decisions are turned into a structured memory so answers stay fast and relevant.
📊
2–3 specialist brains
You get a few focused scorekeepers customized to the problems your AI stack actually has.
🔔
Warnings before drift gets expensive
If a score drops or a threshold is breached, the system warns you instead of waiting for you to discover it the hard way.
📈
A real baseline, not vibes
You leave with a measured starting score and a fixture that can prove whether the brain is improving or drifting.
🗂️
Multi-project awareness
Every project gets a place in the registry, so the brain can answer across your portfolio without losing track of which context belongs where.

Pricing

Start free. Upgrade only if the scan makes you want the always-on version.

Free scan
$0
self-serve

Run the local scan, see whether there is real leakage, and decide from there. Good for anyone who wants a simple answer before talking to a human.

  • Local folder scan
  • Preview brain-health score
  • Plain-English findings
  • No account, no upload, no credit card
Run the free scan
Most popular
Individual / Startup
$100
one-time setup

You are running 1–3 AI projects and want the full always-on version without building the infrastructure yourself.

  • BIV substrate + tick loop deployed
  • Knowledge closets for up to 3 projects
  • 2 specialist brains (LocalLeverage + 1 custom)
  • Push notification warners
  • Baseline eval fixture (≥50 queries)
  • 1-hour walkthrough call
  • You own the code, no ongoing dependency
Email to get started
Enterprise
Custom
scoped per engagement

Multi-team deployments, stakeholder brain networks, custom specialist brains per cost center, and ongoing iteration support.

  • Everything in Individual
  • Unlimited project closets
  • Stakeholder brain network (per-human + per-product)
  • Custom specialist brains per team or cost center
  • Portfolio dashboard + registry API
  • Ongoing iteration sessions
  • SLA and delivery timeline defined upfront
Talk to us

Built by someone running it in production

This isn't a framework. It's the system I run every day across a portfolio of AI products.

JB
John Bradley
johnhavenbradley@gmail.com · BrainOfBrains.ai

I built BrainOfBrains while running ai-cost — a cost-reduction stack for frontier AI. The brain ticks every 30 minutes on my Mac, measures six projects, fires push notifications when something degrades, and has never needed babysitting. I set it up for others because the tooling is the hard part and the value is immediate. If your AI stack costs more than it should and you're not sure exactly where, this is the diagnostic and the fix.

FAQ

Do I need to understand BIV to use this?
No. Think of BIV as the "is my AI setup getting healthier or sloppier?" score. The product explains the score in plain English and tells you what moved it.
Do I need a server or cloud subscription?
No. The tick loop runs as a macOS launchd plist — your machine, your hardware. Specialist brains read from local JSON files. There's no cloud dependency unless you choose to add one.
What AI stack does this work with?
The substrate is stack-agnostic. The specialist brains are wired to whatever cost levers you care about — Anthropic, OpenAI, Ollama local models, hybrid routing. If you have a bridge or proxy, we instrument it. If you don't, we can set one up.
What's BIV exactly?
Brain Information Velocity. A weighted geometric mean of five factors: live token leverage (are you avoiding unnecessary frontier spend?), retrieval quality (is stored context accurate?), compression (how efficiently is context being packed?), freshness (is the brain keeping up with new content?), and self-correction (is the brain calibrated about what it knows?). One number, easy to trend.
How long does setup take?
Target is one business day. We scope your projects, index your content, wire your specialist brains, deploy the tick loop, and run an eval to establish your baseline. You end the day with a live BIV score.
Can a non-technical person use this?
Yes. The easiest path is: run the free scan, read the plain-English report, then book setup if you want the always-on version. The system is technical under the hood, but the experience should not feel technical.
What does "you own the code" mean?
Setup delivers scripts, configs, and closet files to your machine. There's no SaaS layer, no API key to my service, no vendor dependency. If I disappear tomorrow, your brain keeps ticking.
Can I add more specialist brains later?
Yes, and they register themselves automatically. The registry reads any file matching the pattern. Add a new brain, it appears in the next tick. No code changes to the substrate required.

Start with the free scan. Upgrade only if it earns it.

If the scan shows nothing interesting, great. If it shows hidden cost, missing memory, or observability gaps, email me and I will tell you exactly what the full setup would do for your stack.

Email John — johnhavenbradley@gmail.com