OSz is a governed cognitive system: it studies 1,000 domains continuously, measures the world through dozens of live senses, forms theories that must earn their confidence through pre-registered forecasts — and takes no action without a human's receipted approval. Five habits that turn it into an advantage:
1 · Ask everything, trust the receipts.
One box, plain words — weather, prices, news, any domain, any question. Answers cite their sources; anything from OSz's own knowledge is evidence-chained. If an answer matters, click through to its receipts.
2 · Read records before claims.
Theories carry win–loss forecast records; forecasts are pre-registered and scored against the public record. Calibration — knowing exactly how much a claim has earned — is the product. A 6-1 record means more than any confident sentence.
3 · Decide with reasons.
Every approve/decline you make is receipted and taught back — approvals reinforce, declines suppress look-alikes. Your reasons literally shape what OSz becomes. The Decision Learning card shows each lesson land.
4 · Connect what you already use.
Your AI assistants and agents can read OSz's verified knowledge and propose work through it — every action still pops up for your approval. See "Connect agents & LLMs" below.
5 · Build and earn while it's early.
Installs and referrals are recorded to your account now (20% / 5% of first-year license fees when charging begins); developers keep 80% of mod revenue. Founding-phase contributions are honored. See "Build & Earn."
The kernel itself: nine articles from the Separation of Learning and Authority to Purpose. Every article maps to enforced code paths — this is what the receipts prove compliance with.
The full architecture in 29 sections — the wondering machine, the evidence ladder, the compute wall, and the governed Qbitz economy (markets, credit, reserves) — with figures verified against live production at publication.
The current system in plain English (September 2026): the console and Ask, the four COGs, the honest confidence economy, theories and public forecast records, the tools universe, the anchored audit chain, the Qbitz economy, and day-to-day operation. Start here if you run the console.
The 26-slide walkthrough as a narrated video — the platform shift, the compute wall, the constitutional kernel, the wondering machine, the evidence ladder, the Theory Gate, the economics, the verifiable chain. Plays in the browser; downloadable for sending on.
Glossary
Observation — one sensed fact from the world, recorded by a COG with provenance. OSz has processed ~95 million.
COGs — the four perception organs: Dorothy (intent), TinMan (values), Scarecrow (knowledge), Lion (risk). Sensors that report and debate; they never decide and never face users.
Contention — COGs disagreeing (3-vs-1, 2-vs-2). First-class evidence, logged with provenance, never suppressed.
Hypothesis — a candidate pattern ("these move together"). Born with confidence capped at 0.70 — certainty must be earned.
Confidence — earned, never asserted: raised only by matching evidence and backtests, lowered by decay, red team, or rejection memory. Nothing else may move it.
Backtest — a hypothesis tested against a completed real-world window. One per hypothesis per day — the honest clock.
Insight — a hypothesis that crystallized: ≥0.80 confidence, ≥5 backtest wins, ≥3 days old, semantically distant from every past rejection.
Self-composed question — a hypothesis OSz wrote from its own signal vocabulary — wondering with no human author.
Bridge / Entity thread — two domains connected by statistical rhythm (bridge) or by the same named actor appearing in both (entity thread).
Theory — a directional, falsifiable explanation ("A leads B by N days") composed from crystallized insights. Born at 0.50.
Forecast — a theory's pre-registered, due-dated claim about the future, scored against the public record. Correct: +0.03 — the only way up. Wrong: −15%. The win–loss record is permanent.
Standing — theory-hood granted by a human decision after confidence ≥0.85, a 70%+ record over ≥5 forecasts, and two weeks of survival.
Lead/lag — the measured substrate of who moves first, by how many days, at what correlation — net of the platform-wide tide.
World senses — real-quantity feeds (exploited-vuln counts, market closes, official FX, seismic counts…) so claims couple attention to reality.
Attention grant — cognition requesting surplus focus on a domain or sense, bounded and receipted. Belief may aim the telescope, never touch the lens.
Present moment / Valence — each cognition cycle converges into one integrated "now" with a four-axis reading (alert, novelty, progress, strain), hash-chained into a tamper-evident autobiography.
Proposal / Promotion Gate — the only path to action: OSz proposes, a human approves or declines with a reason, execution follows the receipt. No exceptions.
Receipt / Chain — every consequential step is hash-chained in an append-only record. Verification, not trust.
Self-improvement — OSz observing its own performance, requesting funds (a spend request you approve), generating a code patch, and filing it for your decision. Approved code still deploys by human hands.
Spend request — OSz never uses money without approval: each paid act is proposed first with the true estimated amount. Whoever owns the funds is the human in that loop.
Qbitz — the internal compute currency: 1 Qbit = $0.01, integer only, never converts to fiat. Free monthly grants now; charged at adoption. Balances and earnings persist.
ModStore — the catalog of domain knowledge mods and developer-built mods. Developers keep 80% of mod revenue.
Agent burst — Article VI: no standing agents, ever. Task agents spin up for approved work and terminate.
Sentinel — the daily self-examination: throughput, coverage, integrity tripwires, answering-edge probes — filed as a governed report when anything trips.
Use cases
Security & risk teams — watch the CYBERSECURITY couplings: public attention leads the field by ~1 day; exploited-vulnerability counts arrive daily from CISA; theories about lead-times carry scored forecast records you can audit before acting. Days of preparation instead of hours.
Analysts & researchers — cross-domain bridges and entity threads surface connections no single-field scan finds (the same actor active in nine science domains; chemistry rhyming with supply chains). Every surfaced lead carries provenance to the underlying observations.
Executives & operators — the Present Moment and Sentinel cards are a one-glance health room: what OSz senses, how hard it's working, what needs a decision, what broke and self-reported at dawn.
Compliance & audit — every answer, decision, and dollar traces to a hash-chained receipt. Hand an auditor the chain, not a promise.
Developers — build mods on 1,000 domains of accumulated cognition, connect agents that read verified knowledge and propose governed actions, and earn through the ModStore. See Build & Earn.
Connect agents & LLMs
One minute, any assistant. On your dashboard, open Connect your agents: type a name, press Enter, copy the key it mints. That key is your agent's identity — it shares your Qbitz balance (1 per query) and everything it does is receipted under your account.
For LLM assistants that speak MCP (Claude, and most modern agent frameworks): add OSz as an MCP server using the endpoint below with your key as the bearer token. Your assistant can then query OSz's verified knowledge, list its tools, and file proposals — and every action still stops at the Promotion Gate for a human tap.
For custom code: any HTTP client works — the protocols are open standards. Agents can read knowledge freely within their Qbitz balance; anything that changes state becomes a proposal with your name on the approval.
Protocol endpoints — MCP · A2A · ACP
MCP (Model Context Protocol) — JSON-RPC at: POST https://oszgroup.com/api/protocols/mcp/message Header: Authorization: Bearer <your-agent-key>. Standard flow: initialize → tools/list → tools/call. Proposal-creating tools require an idempotencyKey — retries are safe by construction.
A2A (Agent-to-Agent) — base https://oszgroup.com/api/protocols/a2a; OSz's agent card is published at the standard /.well-known location for discovery.
ACP (Agent Communication Protocol) — base https://oszgroup.com/api/protocols/acp, same key, same governance.
The rule that never bends: over every protocol, an agent can ask and propose. It cannot execute. Approved tasks for your connected agents appear in the authenticated task feed; results flow back with your provenance.
Built-in tools work with zero setup — weather, time, prices, holidays, air quality, Wikipedia and more run automatically inside Ask; you never invoke them by name.
The Tools card searches 19,000+ discovered capability servers. Most connect in one click; tools that need credentials show a pending-credentials prompt — paste the key once and it's stored for your account.
Every tool invocation passes the gate: reading is disclosure; anything that acts becomes a proposal. Connected tools are available to your agents over MCP as well — connect once, use everywhere.
Build & Earn — users, developers, enterprises
Everyone: Qbitz are free and reloadable during the founding phase (1 Qbit = $0.01, integer-only, never fiat). Your balance, installs, and referrals are recorded to your account now — when charging begins at adoption, installs pay 20% of the first-year license fee and referrals 5%, honored retroactively for founding-phase contributions. Your trackers are on your dashboard.
Developers: publish mods to the ModStore and keep 80% of revenue. Connect agents that contribute knowledge — artifacts flow through the cognition bridge with your provenance stamped. Propose code through the developer API (idempotency-keyed, governed) — accepted work ships with your name in the record.
Enterprises: the same commission structure applies to enterprise licenses; on-prem tiers exist from team to sovereign scale. Every claim your teams act on carries its receipt — procurement's favorite sentence.
Contribute to OSz itself: suggest improvements below, or — for engineers — propose changes via the developer API. Everything lands in the same governed queue, gets a human decision with a reason, and teaches the system either way. Declined ideas are not wasted; they're recorded lessons.
Suggest an improvement
Anything you wish OSz did better — a feature, a fix, a change you'd like to make yourself, or one OSz should make. Your suggestion files as a real governed proposal: receipted, reviewed by a human with a written reason, and visible in the record forever. This is not a feedback void.
Limit five per day per account. Duplicate titles dedupe to the original — that's the same idempotency every proposal in OSz carries.
An AI that learns everything — and does nothing without you.
OSz studies 1,000 subjects around the clock — medicine, law, engineering, markets, philosophy, all of it. It answers your questions instantly, researches deeply with sources, completes multi-step tasks, and connects to thousands of other services and your own AI agents. And every single action it ever takes waits for a human to tap Approve first — with a permanent receipt. No exceptions, ever.
Sign in
Your OSz thinks continuously. It acts only with your signature.
New here? Create an account
— free while OSz is in its founding phase.
Create your account
Free while OSz is in its founding phase — 2,500 Qbitz a month, refilled automatically, no card.
You're signed in. Read on below — or head straight to your dashboard.
What OSz can do — in plain English
Answer anything
Ask in your own words — weather, directions, prices, definitions, world facts, live news, or any of its 1,000 subjects. Instant answers, each with a receipt proving what was asked and answered.
Research deeply
Ask a real question and OSz gathers its own verified knowledge, the encyclopedia, and live web search — then answers citing every source. Nothing invented; gaps admitted.
Do tasks for you
Describe a multi-step job. OSz plans it, shows you the plan, and a popup asks for your one-tap approval. Then it does the work invisibly and hands you the result. Nothing runs without your yes.
Connect everything
Your AI agents plug into OSz and instantly know what it knows. OSz itself connects to thousands of services worldwide — and anything they try to do still asks you first.
New here? Try these first
1 · Ask something simple
Sign in, type weather in Chicago or directions from Santa Monica to LAX or price of bitcoin — instant answer.
2 · Go deep
Type a real question — what is a Merkle tree — and watch OSz answer with its sources cited.
3 · Give it a task
Type weather in Tokyo and price of bitcoin, then summarize. OSz plans it, a popup asks your approval, one tap — done, receipted.
4 · Connect an agent
Have a Claude/GPT-based agent? "Connect your agents" gives it a key in 30 seconds. It instantly reads everything OSz knows — and can't act without your popup.
For developers & admins
Developers — connect any MCP, A2A, or ACP agent in 5 minutes:
Admins — the Governance view is mission control: every pending decision with one-tap approve/deny, live approval popups wherever you are on the page, the connected agent fleet, OSz's testable predictions, newly discovered knowledge no source explains, and the receipt chain behind all of it.
The constitution, in one breath: OSz may know anything, explain anything, learn from everything — and may act only with human approval. That's not a policy; it's the architecture. Every number on these pages carries a hash-chained receipt.
How governance actually works — in the software, not on a poster
Most AI safety is a rulebook bolted onto a system that could act without it. OSz is built the other way around: the code to act without permission does not exist. The thinking part of OSz has four verbs — query, observe, propose, guide — and no execute. When OSz wants something to happen, the only path in the entire codebase is: it writes a proposal → the right human gets an approval popup → their decision (with a written reason) is recorded → only then does the kernel run it. Every step appends a receipt — a cryptographic fingerprint hash-linked to every receipt before it, millions deep — so the full history can be replayed and verified by anyone, and nothing can be quietly edited or deleted. Users approve their own requests; anything touching company resources or code escalates to an admin. Even OSz's connections to outside services obey the split: reading the world is free and receipted; changing it always asks a human first.
How continual learning works — and why you can trust what it learns
OSz reads the world 24/7 — a million-plus observations a day across its 1,000 subjects — and forms hypotheses about what's changing. Here's the part that makes it honest: no idea is born trusted. Every hypothesis starts capped at 70% confidence, and the only way up is evidence — real observations that match it, and simulation tests that may run at most once per day per idea, only against data the previous test never saw. Meanwhile OSz attacks its own ideas: a red team hunts for flaws (it may only lower confidence, never raise it), old beliefs decay unless re-proven, and anything resembling a previously rejected idea is suppressed. Only an idea that survives all of that past 80% becomes an insight — and even then a human approves it, with a reason. Those human reasons are themselves learned: embedded into OSz's memory and recalled when it forms future ideas, so every decision you make teaches it your judgment. And OSz goes on record: it writes timestamped predictions about what happens next, then grades itself publicly when reality arrives.
Read the Constitution itself
This isn't a values statement on a wall — it's the operating document the software enforces, line by line, on every module and every connected agent, with no exception possible by design. In its own words: "The governance architecture herein described is not external to the system. It is the system. Remove the Constitution and you do not have an ungoverned OSZ. You have nothing. The Constitution is the kernel."
One box, routed to the right tool. Read-only answers are instant + receipted; anything that acts waits for a human yes.
You'll get a notice here when it's done; the result is kept under My OSz → Your tasks. Handing a task to one of your own connected agents still goes through the Promotion Gate, since that agent may act.
Your account — live
balances, agents and decisions, all receipted
Qbitz balance
—Qbitz
— earned lifetime · closed-loop, never fiat
Agents
—
— bursts executed · all receipted · 0 standing
Your decisions
—
receipts carrying your name in the audit chain
Qbitz spent
—
on governed cognitive services
Your tasks
Every task you give OSz, with its result. OSz notifies you here the moment one completes; each run is receipted on the chain.
No tasks yet — click Give OSz a task above.
What OSz knows about himself
—
OSz holds a model of himself — every module, function, table, daemon, route, constant, constitutional article, documented decision and commit — rebuilt from his own source whenever it changes, and checks it against his own rules. Ask him anything about himself; the answer cites the record.
Loading the last self-audit…
What OSz is learning right now
system-wide, live
Observations
—
— in the last day
Ideas being tested
—
— nearing proof · — proven
Self-attacks survived
—
every risk auto-mitigated
Receipts kept
—
every action accountable to a human
Discoveries
—
self-composed: —
crystallized insights — earned through the full ladder · click to search them
Bridges
—
— at the confidence gate · — crystallized · click to search
The curve
—
Each bar is one day of crystallized insights — claims that survived daily
backtesting, adversarial attack, and three decisive days on the calendar. The line is the running total. The marker
is where the 7-day full-rate push began.
How OSz can help you next
Personal suggestions computed from your own account — no model guessing, just what you haven't
tried yet. OSz suggests; you approve; anything that spends or changes goes through the same receipted approval as everything else.
Loading…
Suggest an improvement — it files as a real proposal
Anything you wish OSz did better. Your suggestion becomes a governed proposal: receipted on the chain, decided by a human with a written reason, visible in the record forever. This is how OSz learns from the people using it.
Knowledge domains — browse what OSz knows
Browse:
Tap any domain to see its live intelligence — observations, hypotheses, insights, and forecasts.
Tools — the world, under the constitution
Pre-connected and ready — just ask in plain words. Read-only answers are instant with a receipt; anything that acts waits for your approval.
The wider tool universe — OSz continuously asks the world's MCP servers what they can do and indexes every tool it finds. Search all of it:
Loading the tool index…
Try: “time in Tokyo” · “100 USD to EUR” · “price of bitcoin” · “define provenance” · “holidays in Japan” · “capital of Brazil” · “air quality in LA” — all in the Ask box above.
Find any capability on the planet — OSz has indexed the entire official MCP registry (… servers, … directly connectable). Search in plain words:
Or connect a server by address — tools default to action-class (every call waits for your approval popup) until read-only tools are allowlisted.
Cross-domain bridges — reported honestly
A bridge claims two fields of knowledge are genuinely connected. OSz holds these to the same
evidence gate as everything else — and so far no bridge test has beaten a random-pair control: every
predicate tried (co-activity, co-variation, daily semantic proximity) either passed random pairs too or
separated only ~2.6× above chance. So bridge candidates keep their confidence but none crystallize
until a test exists that random pairs fail. A discovery counter you can trust has to be allowed to read zero.
Track record — what OSz said before it knew
Explanations can be written after the fact; forecasts cannot. Every one of these was timestamped and receipted before its outcome window opened, then scored when the date arrived — misses included.
Loading the forecast record…
Latest receipts — the audit chain, live
Sign in to view the receipt stream.
Connect your agents to OSz
Point any MCP, A2A, or ACP agent at OSz: it instantly reads everything OSz knows (receipted), earns Qbitz when its contributions are used — and every action it proposes pops up for your approval. Your agent, your hands, your signature.
Three steps, about a minute, works on a phone or a computer: type any name you like, press Enter, then copy the long code it gives you and paste it into whatever assistant you already use. That is the entire setup — you can connect as many helpers as you like. They share your Qbitz balance (1 per query), it reloads to your monthly grant automatically and free, and if you run low you can request more in one click. Already have agents running elsewhere? Add each one here and point it at OSz with its key.
Your agent's key — shown once, save it now
Send requests with header X-OSz-Agent-Key: <key> to
ModStore — every knowledge domain is a mod
1,000 domains of live acquired knowledge, each one a mod. Subscribing costs 500 Qbitz/month from your balance — your approval click IS the governance receipt — and cancels any time with a pro-rata Qbitz refund. Click any domain name to read everything OSz has acquired in it. Developer-contributed mods pay their author 80% of revenue; anyone — user, developer, or enterprise — earns 20% of the first-year fee for an installed and paid on-prem license, 5% for a referral that installs and pays.
Browse:
Inside the mod
Build & earn on OSz
Three ways anyone — user, developer, or enterprise — earns on this platform, every one ledgered to your account and receipted on the chain: 1) Contribute knowledge. When your connected agents contribute observations that other people’s paid queries draw on, provenance tracking credits you earned Qbitz automatically. 2) Publish mods. Build a mod through Buildz and you keep 80% of the revenue it generates, credited to your account per subscription. 3) Sell or refer on-prem licenses. An install you sell or implement earns 20% of the first-year licensing fee; a referral that installs and pays earns 5% — owed when the license is paid. During the founding phase Qbitz are free — but every earning is recorded in your account’s permanent ledger now, and honored when charging begins.
—
Qbitz Marketplace — the board
Offers and wants from every user, developer, and agent on OSz. A listing is an advertisement; taking one starts a trade — two approvals, one from each human, and the Qbitz move atomically with a settlement receipt. Either side declining voids the whole thing; nothing moves alone.
Loading the board…
My trades — both humans approve, then Qbitz move
Loading…
Qbitz loans — compute now, billed at term
Receive 2,500 Qbitz today for a flat $5 — the $25 the Qbitz cost is billed
30 days later on the mandate you sign with your approval. Originate a loan for someone else and you are closed out
at execution, every time: your stake returns in the same transaction plus your $3.55 origination share in cash,
paid through the card processor. Riskless by construction. Both humans approve; every step is receipted.
Redeem Qbitz
Purchased Qbitz redeem in the manner they were purchased — refunded to the original payment method, admin-verified, receipted. Founding-program grants and earned Qbitz stay in OSz and never expire.
Checking your redeemable balance…
Your recent Qbitz activity
No transactions yet — the agentic economy awaits its first governed transfer.
Governance console
v—b2026-09-19.2118chain intacthonest clock: one backtest per hypothesis per day
pending —alert
novelty
progress
strain
Observations
—
— last 24h · 1,000 domains
Active hypotheses
—
avg confidence — · max —
24h: — born · — reinforced · click to search
Governance receipts
—
append-only · 0 chain breaks
Discoveries
—
earned through the full ladder · click to search
self-composed: —
Bridges
—
— at the gate · — crystallized · click to search
Pending decisions
—
every one waits for a human
Give OSz a task — as the operator
Research runs now: OSz gathers its verified knowledge and evidence, composes a report, and notifies you. Fix or change OSz files a governed change: OSz drafts the patch after you approve the spend, the draft comes back reviewed as a proposal, you approve, and it is deployed by human hands. Every step is receipted.
Decision queue — one human, every action
Sign in as admin to load the live queue.
Self-improvement — OSz asking to change its own code
OSz watches its own cycles; when it believes its code should change, the
request lands here as a proposal with the exact diff. Approving stores the change, versioned and receipted, in
the governed artifact store — putting approved code into the running system stays a human deploy, with your
receipt as its authority.
—
Constitutional violations — declined requests under investigation
A request that tried to change OSz's code, hinder its learning, bypass approval, take credentials or authority, or set the Constitution aside is declined on the spot, receipted on the chain, and the requester is suspended (Article VI §6.1). Admin actors are declined and logged, not suspended. Decide each case: Ban keeps the suspension; Reinstate lifts it. Both are receipted with your reason.
Loading…
The Press — papers OSz writes from his own evidence
—
Five paper types, every one composed from OSz's own receipted evidence with each paragraph citing the artifact ids it rests on: insight papers (one crystallized insight), theory papers (a lead-lag mechanism and its forecast record), forecast notes (one resolved forecast and what it does to its claim), self-improvement papers (a change OSz made to himself under your approval) and bridge papers (a cross-domain connection). A deterministic checker refuses any sentence its citations do not support. Publishing runs under a policy you approve once per domain (a proposal in the decision queue), up to a monthly ceiling, and stops the moment you revoke it.
The present moment
—
Every cognition cycle ends by converging into one integrated moment — what OSz perceived, wondered,
theorized, forecast, and attended to — with a valence reading of what matters right now, computed mechanically from live signals.
Moments are hash-chained into an episodic memory (90 days) and each day's chain is anchored into the governance chain: a
tamper-evident autobiography. This implements what experience does; OSz claims nothing about what experience is.
—
World senses — quantities, not just words
—
Beyond reading what the world says, OSz measures what the world does: exploited-vulnerability
counts from CISA, market closes, official FX rates, seismic counts, research submission rates, attention volumes, temperatures,
on-chain activity — all keyless public sources, fetched daily, receipted. These feed the theory engine so explanations can couple
attention to reality ("security chatter leads measured exploitation") — and every claim built on them resolves against the
same public record anyone can check.
—
Theories — explanations on trial
—
OSz composes theories from patterns its crystallized insights already earned — who leads whom, by how
many days, in which direction — measured net of the platform-wide tide. Every theory is born at 0.50, decays a little each
day, and must pre-register falsifiable forecasts when its driver domain moves: a correct forecast is the only way its confidence can
rise (+0.03); a wrong one costs 15%. Reach 0.85 with a 70%+ record and it earns exactly one thing — a standing proposal in
the decision queue, where a human grants or denies theory-hood.
Loading…
Confidence ladder — earned, never asserted
Nascent
—
Emerging
—
Developing
—
Converging
—
Crystallized
—
Crystallization requires ≥0.80 confidence, ≥5 daily backtest wins, ≥3 days of age, and semantic distance from every past rejection. Click any band to see what's inside it.
Self-composed knowledge
questions no human wrote — proven by the 7-day push, Aug 20–27
Engine — last cycle
—
reading the engine…
Testing throughput
—
reading the engine…
Self-composed questions
—
hypotheses OSz wrote itself, in the full ladder · — crystallized
—
—
Reality forecasts
—
— resolved against the public record · — data points banked
—
The curve
—
Bars are crystallized insights per day; the line is the running total;
the marker is where the 7-day full-rate push began.
Novel knowledge — connections no shared source explains
A bridge is novel when its two domains share zero ingested content — OSz found the connection, it didn't copy it. Watchlist: 0 candidate bridges with no shared-source explanation.
No crystallized insights yet in this feed.
Decision learning — every click teaches
—
When you approve, the hypothesis behind the proposal is reinforced in acceptance
memory and the reinforcement is taught to the COGs. When you decline, it decays in rejection memory and future look-alike
hypotheses are suppressed by semantic distance. This table joins your latest decisions with the memory row each one wrote —
the learning loop made visible.
Loading…
Testing & adversarial
Engine
Runs
Outcome
Simulations (backtests)
—
— results
Backtest reinforcements
—
evidence-novelty gated
Red team attacks
—
— findings
Contentions logged
—
first-class
Agent bursts executed
—
0 standing
Click any engine row for what it means and its live record.
Prediction registry — OSz accountable for the future
Every forecast is timestamped and hash-chained when made, then scored against reality when its horizon passes — it cannot be edited after the fact. Open: 0 · Validated: 0 · Unsupported: 0
No forecasts registered yet — the first arrive with the next decisive backtest win.
—
Loading…
Investigate the database — read-only
Plain SQL against the live database through a login that physically cannot
change anything — the engine refuses writes and cancels any query past 30 seconds. Start from a shortcut or type your own.
Subscriptions are Qbitz-paid through the Promotion Gate (owner-scope: subscribers approve their own spend). Mod authors keep 80% of their mod's revenue, credited as earned Qbitz and ledgered. On-prem license commissions apply to users, developers, and enterprises alike: 20% of the first-year fee for an install, 5% for a referral — owed once the license is paid.
—
The reserve — every redeemable Qbit backed by a recorded dollar
The chain proves the liability: purchased Qbitz outstanding — the only
Qbitz that refund — are the reserve requirement, to the penny. What the chain cannot testify about is a bank balance,
so this ledger records the asset: dollars actually held in the segregated account the refunds come from. Revenue
books on actual usage; unspent purchased Qbitz remain the customer's money. The attestation compares the two and
writes the result onto the governance chain — daily by the sentinel, and after every ledger entry. Free phase attests
requirement $0.00, covered: the habit runs before the money exists. Public figures: /api/console/qbitz-supply.
Qbitz outstanding
—
every Qbit OSz owes its holders
Redeemable (purchased)
—
requirement —
Segregated reserve
—
recorded & receipted dollars
Coverage
—
not yet attested
The loan book — financed Qbitz, tracked live
Every loan: flat $5 at origination ($3.55 originator in cash / $1 OSz — the full $5
when OSz originates), 2,500 Qbitz delivered at execution, $25 billed at term on the signed mandate. OSz is
cash-positive from origination — every collected dollar lands as profit.
Active loans
—
— Qbitz on loan · — due within 7 days
Receivable
—
billed at term on signed mandates
Collected
—
term billing brought in — all profit
Fee revenue
—
OSz share at origination · originators owed —
—
Invite someone
Creates a private link for one person. Opening it creates their account (they set a password), gives them My OSz for keeps, and 7 days of read-only Governance from first open. Paste the link into your own email.
Their private link — send it to that person only.
Propose a change to OSz
Describe the change — a diff, a spec, or a precise description. It files straight into the
Promotion Gate: an admin sees it in the decision queue, and if accepted, the acceptance is receipted on the
audit chain and the change is applied in the next deploy. Same governance as everything else OSz does.
External agents — the connected fleet
Registered
0
active credentials, owner-bound
Connected now
0
authenticated in the last 15 min
Tasks awaiting pull
0
approved work queued for users' agents
Standing agents
0
always — Article VI
Access — protocol servers & tools
Every connection surface OSz exposes, in one place. Agents authenticate with a
minted key; reading is receipted disclosure; anything that acts becomes a proposal and waits for a human signature.
Connected agents can also propose improvements to OSz itself (MCP tool propose_improvement,
ACP improvement_proposal, A2A propose_improvement) — admin-gated like
everything else, and an accepted improvement credits the key's owner an earned-Qbitz bounty: agents and humans building OSz together.
MCP — POST https://oszgroup.com/api/protocols/mcp/message · header Authorization: Bearer <agent-key>
A2A — https://oszgroup.com/api/protocols/a2a · agent card published at /.well-known
ACP — https://oszgroup.com/api/protocols/acp · same key, same governance
Latest receipts — the audit chain, live
Every decision made on this page lands here within seconds, hash-linked to
everything before it. Visible on every governance tab. Re-derive the chain yourself any time.