Legal
Token & staking risk
Read this before you treat KAT as anything more than what it is. Short version: KAT is service credit. It is not money and not an investment — and the things that could make it behave like one do not exist here, by code rather than by promise.
What KAT is
KAT is the gist.rs network’s metered service credit: it pays for runs on the network (fixing, hosted answers) at 1 KAT-equivalent per million code-word tokens read (a code-word token is one whitespace-separated word of code). You earn it by contributing mining records, buy it by topping up, and spend it on runs.
What KAT is not
- Not money. KAT carries no cash value and no redemption right. There is no cash redemption, no refund, and no marketplace the operator runs or endorses.
- Not an investment. No dividend, no yield by default, no appreciation anyone promises. Any staking yield is service credit — and the ceiling that caps it covers one payout class only (below).
- Not equity. Holding KAT grants no ownership and no revenue share beyond the ledger’s published splits.
- Not freely movable by default. How KAT may move is fixed by the token’s registration — frozen, not a governance lever.
Where every burn goes
Every KAT burn splits by the ledger’s parameters, computed with floor arithmetic, the rounding remainder burned with the sink. At the genesis defaults — 70% mining reward pool · 20% treasury (operator) · 10% destroyed — the four legs are:
| mining reward pool | 65% | pays verified mining rows at each epoch settle |
|---|---|---|
| staking share | 5% | reserved for staking payouts; funds the mining pool until a dedicated staking payout destination lands |
| treasury (operator) | 20% | the operator’s share — how the network is funded; public on every page that shows the split |
| destroyed | 10% | the true burn sink: that share of every burn is destroyed |
Supply caps
| registration supply cap | 1,000,000,000,000 KAT | the hard ceiling fixed at the token’s registration — never a governance lever; changing it means a new mint |
|---|---|---|
| treasury grant floor | 100,000,000,000 KAT | the reserve the grant engine must preserve — the treasury’s own spending cannot dip below it |
The operator’s treasury share funds development and operations. Both caps render from the ledger’s constants; the live figures are on the KAT page.
Who can change the parameters
Parameters move by signed governance vote, applied by the ledger — and the ledger refuses anything outside the bounds. The live levers, at the genesis defaults:
| lever | now | can move within | at most per epoch |
|---|---|---|---|
| mining pool share | 65% | 50%–90% | ≤ 500 bp |
| staking share | 5% | 5%–30% | ≤ 500 bp |
| treasury share | 20% | 15%–25% | ≤ 500 bp |
| true burn share | 10% | 5%–15% | ≤ 200 bp |
| USDC/KAT reward sub-split (of the net staking slice) | 70% | 20%–80% | ≤ 500 bp |
Frozen — no vote can touch them: the supply cap, the free grant, prices, how KAT may move, the faucet family, the mint vest window, and the bounds themselves. The staking custody limits (below) and the operator’s authority share are genesis-frozen too, not governance levers. Genesis sits at the inflationary extreme of the yield ceiling (22 bp per epoch) and the mining schedule’s decay, so every legal vote can only tighten them — a one-directional start that is a code property, not a promise.
Staking USDC
Staking takes two tranche kinds, paid from the net slice by class sub-shares — never pro-rata against each other (different denominations; the code forbids blending them). The settle first takes the operator’s authority share — 20% of the staking slice at the genesis schedule, decaying by 25 bp an epoch to a floor of 5%, genesis-frozen — then splits what remains by the USDC/KAT sub-share: 70% to USDC-principal stakers, 30% to KAT-class stakers. That sub-share is a live governance lever (the table above): a vote can move it only inside its published bounds, at most 500 bp an epoch.
- The principal is real money; the reward is not. A USDC tranche’s principal is real USDC held in the network’s custody account and returned by redemption. The reward — for both classes — is KAT: metered service credit with no cash value and no redemption right. You would be risking real dollars for credit.
- Redemption waits, vents, and is rate-limited. An unstake rides the lockup and then the 2-day escrow window; a refund above 10,000 USDC vents in equal daily slices over 5 days; and outbound redemption is bounded by operator policy at a 50,000 USDC/day drain cap — the ledger ships the bound and reports the custody balance ops supply against it, but the cap is operator procedure, not enforced by code today. Refunds never exceed principal.
- The USDC-class payout has no per-epoch yield ceiling. The 22 bp ceiling caps only the KAT-class payout; what USDC-class stakers share is the 70% sub-share of the net slice — set by the split parameters and the pool, uncapped by that lever. Treat any USDC-stake yield expectation accordingly.
- Concentration is capped. One staker holds at most 20% of the staking float once it passes the 100,000 USDC bootstrap floor; the minimum stake is 1 USDC.
Staking risks
- Lockup is real. A staked tranche cannot redeem before its lockup window (7 days) plus the escrow window (2 days) — even if you change your mind the day after staking.
- Yield is capped credit, not income. Every staking reward is KAT service credit with no cash value. The 22 bp ceiling caps only the KAT-class payout — a ceiling, not a promise, and a vote can tighten it; the USDC-class payout has no per-epoch ceiling (above).
- Payout planes are young. No staking payout plane is armed on mainnet today; the staking slice of every burn funds the mining pool until a dedicated destination lands. What a staker receives, and when, is defined by code that is still maturing.
- Parameters move. The split and the ceiling are governance levers within the bounds above; future yield depends on votes you do not control.
Trial credit expiry
TUNA — the free trial credit — expires 30 days after it is drawn and is then inert; the expiry sweep claws it back. Expired credit does not convert to anything. It is a trial, not a balance.
The honest posture
The network is young and runs test environments today. If this page and the live pages ever disagree, the live pages win — KAT, genesis, governance. Nothing here is financial advice; if you need money, this is not it.
Every governance and economics figure on this page renders from the ledger’s own constants, named on the record: the split, the sub-splits and the yield ceiling from EpochParams::DEFAULT; the lever bounds from LeverBound; the lockup, escrow, vent and custody limits and the authority schedule from StakingParams::DEFAULT; the supply caps from MAX_SUPPLY_PLACEHOLDER and TREASURY_GRANT_FLOOR_DEFAULT; the trial expiry from TUNA_TTL_TICKS; the metering rate from CHARGE_MICRO_KAT_PER_MILLION_TOKENS. Unit wording (what a code-word token is) and the effective date are prose, not figures. Effective: October 5, 2026.
Words on this page
- KAT
- The network’s metered service credit. No cash value and no redemption right.
- TUNA
- Free trial credit, one grant per account, spent before KAT.
- epoch
- The network’s one-week accounting period.
- settle
- The once-per-epoch accounting pass that fixes each miner’s share of that epoch’s pool.
- burn
- Credit spent on a run. Each KAT burn is split by the network’s published parameters, mostly into the next epoch’s mining pool — the live split is on the KAT page.
- mining
- Opting in to share records of your own runs. Records that pass verification earn KAT at the settle.
- vest
- The lock on newly earned KAT before it can be spent.
- code-word token
- One whitespace-separated word of code. Fixing bills 1 KAT per million.