Why Omnial exists
Your AI is limited by what it can reach.
It can talk, draft, analyse, plan. But the moment a task needs something real: live data, a scraped page, an image, a verified contact, it stops. Every one of those capabilities lives behind a vendor's account, a key, a bill and a contract nobody wants to babysit. This is the door we built instead: one connection, one balance, and your AI does the reaching.
The wall
Every capability your AI wants is locked behind its own door.
Not one hard wall. A thousand small ones, each run by a different company with its own rules.
A new account for every job
Search is one vendor. Enrichment is another. Media is another. Each wants its own signup, its own API key, its own bill and its own dashboard. Your AI does not care which one it is, but you end up managing all of them.
A schema is not a teacher
A tool's input schema tells an AI which fields exist. It does not tell it which phrasing earns the good result, which option costs extra, or what fails quietly. An AI handed a raw spec uses the tool badly and still gets the bill.
Cost is a mystery until after
Most tools price you after the fact, or not at all. Your AI cannot plan around a price it cannot see, and you cannot budget against a number you only find on the invoice.
The door
One connection. One balance. The rest is your AI's job.
Connect it once, and the loop your AI runs looks like this. Only the last step spends anything.
Say what you need
Your AI asks for the capability in plain words, like “find this company's open roles” or “scrape this page”, and gets ranked matches, each one carrying the reason it matched. If nothing matches, it says so instead of guessing.
It reads the usage guide
Before it spends a cent, it reads the usage guide for that exact tool: what each option does, the patterns that work, what drives the price. A schema tells it the shape; the guide teaches it the craft.
It prices the call, then makes it
It can price the exact input it is about to send and spend nothing. When it runs the real thing, the call runs behind a hold on your balance. A run that uses more than quoted can be charged up to twice its quote. That can take your balance below zero; new runs are refused until you top up.
It pays for that one call
The actual cost settles out of your balance and any remainder of the hold is released. A call that failed is not charged for. You pay per use, not for a subscription.
Most calls cost a small fraction of a dollar. The unit of purchase is the thing your AI actually did, and the price shown is the price you pay.
No monthly minimum. Top-ups start at $5. No per-vendor accounts and no key per provider. Your AI finds the one tool it needs, when it needs it.
Who it is for
Built for the person, the builder and the business.
The same door works differently for each of you.
The business owner
You get an AI that does, not just advises. It researches, pulls the data, drafts the asset, checks the number, and hands you the result. You don't sign up for ten services to get one job done.
The developer
One surface your AI actually uses well: a clear contract on every tool, a price for the exact input before it runs, a usage guide that teaches it, and a balance you can meter. Fewer keys, fewer dashboards, fewer surprise invoices.
The agency and the builder
Put the capability inside your product or behind your own agent. Your users get the outcome, and you do not have to stand up, pay for and babysit a stack of upstream subscriptions to get it.
Why you can trust it with real money
The decisions that make it safe to hand to an AI.
An AI with a balance can spend money. These are the six decisions that keep that safe, with the technical detail underneath.
It knows what it will get back before it pays
Every tool declares not only what it takes but what it gives back. Your AI sees the shape of the answer before it commits, so it never learns the format by paying for a call and looking.
omnial_inspect returns the input schema and the output schema together, with the pricing model, a cost estimate for your exact input, measured latency and success rate, and the rate limits your key is held to. Declaring an output schema is a condition of shipping at all.
Money is exact, all the way down
Nothing near a balance is a floating-point number. A real rate card is quoted at a thousandth of a cent, and a system that rounds it into a float quietly accumulates a gap between what it charged and what it says it charged.
Balances are integer nanodollars (one in a billion of a dollar) on the wire, in the database and in every step of the arithmetic. The ledger is append-only, with a full balance snapshot on each entry, so any balance can be re-derived from its own history.
A price is a price
Real pricing is conditional: a base that depends on the mode, an extra past a threshold, a separate charge per page. A price that has to be flattened to fit is a price somebody is wrong about.
Prices can depend on the input, like “more than ten results”; every condition is shown, and a composite price is a sum over exactly the components that apply to that call. The catalog build proves those components are mutually exclusive and refuses a price that knowingly omits a real charge.
Your AI can stop a run and find one it lost
Two failures that cost real money: a long job started with the wrong input, and a crashed process that no longer holds the handle to a run still holding its money. Both have an answer.
omnial_cancel asks the provider to stop first, then settles and releases what is left of the hold, in that order. omnial_runs lets a restarted agent list its own in-flight runs. Where a tool cannot be stopped, the system says so instead of pretending.
No confident garbage
When a tool can't do the job, you get a clear ‘no’ and the reason, not a guess you pay for.
Search always answers, ranked, and every hit carries the reason it matched, including a note when nothing matched and this is the nearest entry. The same ranking serves this website's own search box, so what you can check here is what your agent sees.
A guide, not just a spec
Every tool has a usage guide. A schema tells an AI which fields exist; it does not tell it that a scraper bills per result, so one query at a time is cheaper.
Each tool ships a usage.md written for that tool and checked against its schema: what the parameters do and how they interact, the patterns that work, the failure modes to expect, and what drives the price. A tool cannot ship without one, and it cannot ship a worked example that does not validate against its own input schema. Your AI receives the full document at runtime, byte for byte.
Where we are today
What is live, and what is not built yet.
The catalog is live. Here is the part that is still missing.
Live and growing
Real, working integrations are browsable and callable now, with live pricing and settlement. The catalog keeps growing as more integrations go live.
No long-lived resources yet
Every call here is request-shaped: a run holds money, executes and settles. Tools that hand back something that outlives a call, like a browser session, a sandbox or a rented proxy, are not built yet.
Building on Omnial? Calling tools from your product is fine; reselling the catalog isn't. Details
Check any of it yourself.
Every tool page publishes the full contract and the pricing model, itemised charge by charge. Your agent receives the usage guide separately, at runtime. The install is a file you hand to your AI. It does the rest and comes back to tell you it is ready.
Questions? Write to pixellabsweb3@gmail.com.
Live catalog
Tools are live and callable. What that means