AI opportunity infrastructure / in development
AI-powered opportunity evaluation before real capital is deployed.
We read opportunity signals, score them transparently, and pressure-test the economics on paper — so a launch is only ever considered after the evidence, never before it.
01 the problem
Signal moves faster than analysis. Momentum, audience and narrative are visible immediately; creation cost, fee share, organic traction and operating drag are not.
By the time those numbers are modelled, capital is often already committed — and there is no durable record of why the decision was made.
A signal that spreads is not the same as an opportunity that works.
Launch cost, gas and operating overhead rarely appear in the original pitch.
Without a persisted verdict and reason code, failures are only felt, never analysed.
Fomo Alpha Labs exists to evaluate the evidence before capital is committed.
02 how it works
A public opportunity signal is validated, link-checked and stored as a paper-only record.
Six weighted dimensions produce a transparent score against a fixed, published pass threshold.
Exactly one simulated experiment runs per passing candidate, on a bounded clock.
Simulated volume becomes a creator fee; launch, gas and operating cost are deducted to give net P&L and ROI.
KEEP only with minimum traction and non-negative economics. Otherwise ABANDON, with a recorded reason.
Evidence is handed to a fail-closed readiness gate. It never starts anything on its own.
03 ai opportunity engine
Every candidate begins as raw text — a public post or opportunity note. The engine decomposes it into named dimensions, returns a score, and returns the reasoning alongside the number.
Current published heuristic weights. Output is reproducible and auditable.
End-to-end example SIMULATED
Illustrative path for one candidate. Example values only — not a recorded run, and reaching readiness does not start live trading.
04 paper experiment
One experiment per passing candidate. Simulated buyers and volume produce a creator fee; launch, gas and duration-based operating cost are deducted; the result is net P&L and ROI. Every figure is simulated — never on-chain, never a real balance.
05 why paper first
Every idea runs as a paper experiment. Experiments generate simulated activity, apply the real cost structure, and persist a verdict with a reason code. Only when the economics hold up does an opportunity move forward at all. Capital is the last step in the sequence — never the first.
06 decision layer
Both outcomes are persisted with an explicit reason code, so every result can be audited and compared later.
Proceed to evidence review
KEEP means the hypothesis earned a longer look — nothing more.
Stop, and keep the learning
Abandoning cheaply and on purpose is the point of paper experimentation.
Traction alone can never produce a KEEP. A losing experiment is always abandoned, however many simulated buyers it attracted.
07 evidence history
Persisted paper experiments, read from the local ledger export. Each row carries its own verdict and reason code. Every figure is simulated — never on-chain, never a real balance.
No experiment records in this snapshot.
Records appear here once a paper experiment reaches a KEEP or ABANDON verdict. Nothing on this page is generated for display.
Evidence snapshot could not be loaded.
The static file evidence.json was not reachable. When it loads, only
persisted records are shown — never generated ones.
08 launch readiness
A KEEP result moves an opportunity toward launch-readiness review. It does not deploy anything. Between the two sits an explicit gate that is designed to fail closed.
Gate remains fail-closed until every prerequisite is deliberately supplied by a human.
Current state: 7 checks ready, 3 configuration prerequisites open. The open items are prerequisites — not errors.
Configuration prerequisites — not errors.
Nothing on this list is filled in automatically. No address is supplied, no key is read, and no transaction is possible until a person supplies each prerequisite and authorises the step.
09 current status
Pre-production. The lists below reflect the actual state of the system — no users, revenue, funding, partnerships or profitability are claimed anywhere on this page.
Next milestone
Passing paper economics does not automatically enable live trading. Each step is a separate, deliberate action, and the last one only happens with an explicit human authorization.
All paper activity is simulated. Simulated buyers, volume and fees do not represent real market participants, real funds or real payouts.
10 architecture
Each layer has one job and hands a durable record to the next. Nothing crosses a layer boundary implicitly.
Validated text and link intake, versioned signal records, published scoring threshold.
Dimension decomposition, weighted scoring, PASS / REJECT with returned reasoning.
One experiment per candidate, seeded simulated activity, append-only ledgers, atomic state.
Creator fee, launch and gas cost, duration-based operating cost, net P&L and ROI.
KEEP / ABANDON evaluation with persisted reason codes and full decision history.
Fail-closed gate over routine, parameters, balances, monitoring, exclusion and idempotency.
State lives in plain, inspectable files on the machine that runs it. No hidden database, no opaque queue — easy to audit, easy to reset.
The observation path can read chain state but has no code path to sign or broadcast. That separation is enforced by tests, not by convention.
This site is plain HTML, CSS and JavaScript — no build step, no framework, no API keys, no backend dependency.