⚡cyber/$CYB.md

root token of cyber. unit of stake, fees, and rewards for proven contribution to focus φ*. emission schedule is a function of time alone; who receives emission is a function of φ*. see whitepaper, mining, staking

role

Focus is the scarce object: the unique fixed point of the tri-kernel over the cybergraph. Moving and creating focus costs work. CYB makes that work transferable — spend to link and prove, earn when your links raise collective focus (proven Δφ*). It is not a fee token bolted onto ranking; ranking and payment share one physics

hard money

$CYB is hard money in the oldest sense — expensive to make, impossible to counterfeit — and in four new ones:

• backed by conservation. focus sums to one across the whole graph — the backing quantity cannot be printed, only redistributed. Emission happens only against proven Δφ*: new supply exists exactly when the graph provably grew smarter. Inflation is evidence of knowledge, not a policy
• quantum-hard by construction. Identity is a hemera hash (Poseidon2 over the Goldilocks field), state is a polynomial commitment, proofs are zheng — hash-and-field cryptography with no discrete log for Shor to unwind; key exchange in the mudra ladder is isogeny-based (dCTIDH)
• private by the same field. The field that proves also encrypts: strata hosts the FHE regime (jali R_q, TFHE in the mudra ladder), so balances can compute without being seen — prove the property, keep the number
• it means something, literally. Every unit traces to links in the cybergraph, and neural reads value as position in the topology. This is money whose backing you can query

And the mint is in your pocket. Proving work targets the devices people already own — Mac and Android — with direct access to the silicon through honeycrisp: NEON/AMX/SME, Metal GPU, ANE, zero-copy unified memory. A million laptops and phones instead of three datacenters: issuance lands where the hands are

cap

Total supply is the order of the Goldilocks field used by nox and proofs:

p = 2⁶⁴ − 2³² + 1 = 18,446,744,069,414,584,321

Cap is arithmetic, not a governance vote. On the bootloader (bostrom) the same energy currently circulates as $C

genesis

At block 0, $C holders receive 187,416,084,623,451,570 CYB ≈ 1% of p (281,405,532,467,645 snapshot × 666). Continuity of prior stake into the soft3 field; remaining mass is reserved for the emission schedule

emission

How much exists at network age t (years) is fixed by the clock — identical on every honest node, no oracle, no forgery surface:

M(t) = p · (1 − (1 + t/τ)^(−k)), τ = 0.33 year, k = 0.5

π(t) = M′(t)/M(t) is instantaneous inflation (1/year). There is no discrete year step and no halving epoch

why this shape

The graph needs proving and settlement capacity most when it is empty and growing. A power-law head puts most of the emission in the first years so early useful work (mining division and fold, early links that set topology) is paid while the network is still cheap to dominate with capital alone. Finite initial rate k/τ ≈ 152% of cap per year avoids a single-block flood; the year-1 integral still reaches ≈ half of p, so the main prize for compute is front-loaded without a cliff

The polynomial tail keeps residual issuance for centuries under the same cap: long-run security budget without reintroducing a policy rate. Scale-free graphs and Zipf focus already are power laws; emission matches the same family so the money supply does not impose a foreign timescale (e.g. fixed halvings) on a structure that does not have one

Clock and focus stay separated: M(t) only answers how much is available; stake-weighted Δφ* answers who mints (see allocation)

schedule

revenue

CYB is a coin under tok / plumb. Four ops: pay, lock, mint, burn. Σ balances = mints − burns

Emission $M(t)$ bootstraps the field when it is empty. Revenue is the long-run metabolism: every movement of energy pays a tax on diffusion. That tax is the main source of self-development once the schedule thins — same merit brain as inflation, different faucet

tax on diffusion

Every transfer of amount $G$ that moves CYB — pay, lock, unlock, including staking — pays $\tau = 1\%$:

$$ \text{received or locked} = (1 - \tau)\,G, \qquad V = \tau\,G $$

No carve-out for staking: locking mass onto a claim is still diffusion. Hopping energy without improving the graph is not free

how $V$ splits — half burn, half to everyone

Simple rule on the tax pot $V$:

$$ \text{burn} = \tfrac12\,\tau\,G, \qquad B_V = \tfrac12\,\tau\,G \ \xrightarrow{\ \text{hybrid }\alpha\ }\ R_{\mathrm{PoW}},\ R_{\mathrm{PoS}} $$

So: half destroy, half pay everyone who earns — “everyone” means every channel that already earns under emission, not a pro-rata airdrop to all holders. Creator residuals (creator_mint_share, creator_pay_share) still skim their frozen shares of mints and service legs through robots; they do not take a fixed half of $\tau$ off the top

pay / lock / unlock  →  τ = 1%  →  ½ burn  +  ½ → B_V → same hybrid as emission
M(t), floor, service →  B     → hybrid α → PoW / PoS / Δφ* mint

Early life: $M(t)$ dominates. Mature life: $B_V$ from velocity dominates self-development. One merit function, two faucets

network effects

Not a checklist of actions — seven factors that compound. Each one makes the next stronger; $CYB growth only where knowledge got better

fairness  +  finality  +  truth
        → intelligence
          → revenue
            → efficiency
              → population
                → fairness … (again, larger)

Eighth factor if you stretch the ring: karma — quality of actors so population is not spam. Kept off the wheel for now; it multiplies truth and intelligence rather than standing alone

Hover a factor.

mint

Coin. Two faucets, one hybrid brain under adaptive hybrid economics (PID on $\alpha$, floor, $\beta$):

• schedule envelope + security floor + optional service fees $\to$ pot $B$
• half of diffusion tax $\to B_V$

$$ R_{\mathrm{PoW}} = B_{\mathrm{tot}}\,(1 - \theta^{\alpha}), \qquad R_{\mathrm{PoS}} = B_{\mathrm{tot}}\,\theta^{\alpha}, \qquad B_{\mathrm{tot}} = B + B_V $$

Δφ* knowledge mint only if proven $\Delta\phi^* > 0$. Security floor mints only to PoW + active stake; PID-decays as velocity ($B_V$) and fees cover security. Passive lock ($v = 0$): rank only. Spec: rewards

Card. mint robot/neuron = identity Card, not CYB. Create may cost pay/burn. At create freeze creator_mint_share, creator_pay_share $\in [0,1]$: residual on later Coin mints and service-leg pays through that agent — including when the agent earns from the “to everyone” half. Optional royalty on card transfer. Not a second $M(t)$

M(t), floor, service  → B  ─┐
pay/lock/unlock τ=1%  → ½ V → B_V ─┼→ hybrid α → PoW / PoS  [− creator shares]
                      → ½ V → burn  ┘

lock

Freeze CYB on a cyberlink (staking, will). Lock and unlock are transfers under the tax on diffusion (½ burn, ½ into $B_V$) — staking is not a tax-free parking lot. Of the locked mass after tax, active ($v = \pm 1$): weight + stake mint + share of reward pots × karma. Passive ($v = 0$): φ* influence only. Wrong active bets lose score under BTS; idle capital does not compound mint

pay

Pay, lock, unlock — all move $G$ under $\tau = 1\%$ (half burn, half $B_V$). Headers, queries, DA, inference are ordinary pays under the same $\tau$; the peer may still receive the service leg after tax. creator_pay_share skims service residual through robots, orthogonal to the half that entered $B_V$ for everyone

burn

Half of every diffusion tax ($\tfrac12\tau G$) is destroyed — discipline of hoppers, deflation under high velocity. Also: eternal particle / cyberlink (φ* floor); optional slash-to-burn. Mint makes energy transferable; burn makes empty circulation expensive and deliberate eternal influence permanent

allocation

How much: $M(t)$ + floor + $B_V$ (half of velocity tax) + service fees. Who: Δφ* Shapley value · fold work · active stake × karma. Zero stake and zero work → zero. The other half of velocity tax burns. rewards · adaptive hybrid economics · tok · self