Legaicy Whitepaper

Executive Summary

Legaicy is a crypto-native network for paid human insight. Users answer questions, build trust, and earn LGCY for high-quality work. Corps pay ETH when that work is selected for funded research, either from the existing corpus through Path A rotation or through fresh panels.

The economic split matters. LGCY is emitted for work quality and auto-staked into the answered tier. ETH is corp demand expressed in money. That makes the corpus an accumulating labor surface rather than a one-shot answer sink.

A clean corpus answer earns LGCY at submission and can keep earning ETH later through Path A. But it does not sit at the front forever. Path A uses a per-question FIFO rotation queue inside the top 10th percentile of qualifying answers, so recurring income stays real without letting the same old winners capture every later query.

Token

1,000,000,000 LGCY hard cap with a 5,000,000 pre-mint and halvings at 100M, 250M, and 500M. Past 500M, emissions follow an asymptotic decay curve — total supply approaches 1B asymptotically; the cap is enforced as a hard stop (post-cap, new emission halts; existing balances continue to vest and remain withdrawable).

Corp-ETH gate

ETH-eligible users must clear tier stake thresholds of 10K / 25K / 50K / 100K LGCY in C / B / A / S.

Distribution rule

Corp ETH goes only to the top-N answers by quality, with proportional share inside the cutoff.

The Opportunity

The open internet is filling with synthetic text, but useful human judgment is not scaling with it. That creates a gap between content abundance and trustworthy signal.

Legaicy is built for that gap. It does not try to make every answer valuable. It tries to make high-quality, trust-bearing answers searchable, selectable, and payable.

The Product Model

Users

Users subscribe, answer corpus questions, accumulate trust, earn LGCY into tier buckets, and can later receive ETH when corps select their work.

Research buyers

Buyers search the corpus semantically, add filters, price a query, and then fund either a corpus retrieval path or a fresh active panel.

User Journey

Users move from the public site to the public dapp landing, then through wallet connection, onboarding, subscription, and finally into the connected workspace.

Flow 1 is the corpus-answering loop: answer a question, clear the quality floor, and earn LGCY auto-staked into the answered tier. Flow 2 is the corp-jobs loop: browse or be matched to funded work and compete for ETH distribution.

Earnings Model

LGCY and ETH are paid for different things.

LGCY is paid for high-quality answers and auto-staked into the relevant corpus tier.
ETH is paid by corps when those answers are accepted for funded research.

Locked emission formula LGCY = base × (quality / 100) × trustMult × demoMult × tierMult × verbatimMult × (1 + bonus_rate / 100)

Factor	Locked behavior
`base`	Starts at 100 LGCY before halving effects.
`quality`	Must clear 60 / 100 to earn anything.
`trustMult`	`1.0 + (TrustLevel - 1) × 0.1`, from 1.0x at T1 to 1.9x at T10.
`demoMult`	`1 + (filledPct / 100) × 0.5`, capped at 1.5x, active only at Trust 2+.
`tierMult`	C = 1.0, B = 1.5, A = 2.0, S = 2.5.
`verbatimMult`	1.0 by default, 1.25 when that answer was verbatim-consented.

Earnings Examples

The formula becomes easier to understand when it looks like a character build instead of a spreadsheet. The examples below all assume base = 100, quality = 85, and a clean answer above the quality floor.

Example 1: Fresh subscriber

Trust 1, no demographics filled, no verbatim, near-zero EXP. This is the first-stage build: subscribed, answering regularly, and still living mostly in the low-tier draw pool.

Tier C question
100 x 0.85 x 1.0 x 1.0 x 1.0 x 1.0 x 1.0 = 85 LGCY

Tier S question
100 x 0.85 x 1.0 x 1.0 x 2.5 x 1.0 x 1.0 = 212.5 LGCY

Rounded: about 213 LGCY.

A conservative three-answer day lands around 280 LGCY/day, with lucky S-tier hits pushing the total higher.

Example 2: Mid-progression user

Trust 5, full demographics, verbatim enabled, around 500,000 EXP for about a 20% bonus. This is the user who has put in real reps and started stacking meaningful multipliers.

Tier B question
100 x 0.85 x 1.4 x 1.5 x 1.5 x 1.25 x 1.2 = 401.625 LGCY

Rounded: about 402 LGCY.

Tier S question
100 x 0.85 x 1.4 x 1.5 x 2.5 x 1.25 x 1.2 = 669.375 LGCY

Rounded: about 669 LGCY.

A strong three-answer day lands around 1,000 LGCY/day, with better draw mixes climbing above that.

Example 3: Veteran build

Trust 9, full demographics, verbatim enabled, around 5,000,000 EXP for roughly a 96% bonus. This is the late-game build where high-tier draws are the normal case instead of a rare outlier.

Tier S question
100 x 0.85 x 1.8 x 1.5 x 2.5 x 1.25 x 1.9615 = 1407.2 LGCY

Rounded: about 1,407 LGCY.

Tier C question
100 x 0.85 x 1.8 x 1.5 x 1.0 x 1.25 x 1.9615 = 562.88 LGCY

Rounded: about 563 LGCY.

A veteran-weighted three-answer day lands around 3,800 LGCY/day because S-tier becomes the dominant draw instead of the rare upside.

These are per-answer examples. The point is not that every day looks identical. The point is that the multipliers stack in a way that rewards consistent, high-quality, filled-out, trust-building participation, and that the strongest builds are visibly stronger than fresh accounts.

Trust Progression Formula

Trust progression is locked as hybrid clean days plus clean answers. Promotion from level N to N+1 requires N × 10 clean days and N × 10 clean answers.

There is no trust decay and no trust demotion. The trust ladder is meant to reward consistency, not churn users through resets.

EXP and Bonus Emission

EXP equals lifetime LGCY earned on a 1:1 basis. It is soulbound by convention and feeds the bonus multiplier:

Bonus formula bonus_rate = 100 × EXP² / (EXP² + 10^12)

Reference points remain: 1M EXP = ~50%, 2M = ~80%, 3M = ~90%, 10M = ~99%.

Per-Tier Vesting

LGCY is auto-staked into one of four corpus tiers and vests on four independent clocks: C, B, A, and S.

1% per clean tier-day
100% cap per tier
24-hour claim cooldown per tier (independent — claiming from one tier does not block claims from another)

Corp-Side Experience

The corp flow now starts with AI semantic search, not a rigid pricing menu. Buyers describe what they want, the system searches the corpus, branches into existing or custom questions, layers filters, runs feasibility checks, prices the request, and then routes into fulfillment.

Buyers care about filters, not static packages. They want the right number of completes, the right audience, the right response format, and a clear line between corpus matches and fresh panels.

Fulfillment Paths: Path A and Path B

Corp queries now resolve through two fulfillment paths.

Path A uses a per-question rotation queue over the top 10th percentile of qualifying answers. Path B opens a fresh panel when there are not enough qualifying answers or when the corp needs custom questions.

Path A

No panel opens. Users are picked from a per-question FIFO rotation queue built from the top 10th percentile of qualifying answers. Once paid, they drop to the bottom. If their answer no longer clears the current percentile bar, it ages out until they re-answer and re-qualify.

Path B

A live panel opens on the job board. Eligible users claim slots first-come, first-served and compete on answer quality for ETH distribution.

Legaicy is permissionless on the buyer side. Any party with ETH can fund a query through the on-chain payment flow. Legaicy does not pre-approve or vet buyers; instead, the Buyer Terms of Service restrict how the data may be used.

Filter-Based Corp Pricing

The pricing model is lane base x (1 + additive premiums) x completes.

Lane or filter	Locked price / premium
Pulse lane	$2.50 / complete
Targeted lane	$8.00 / complete
Verified lane	$30.00 / complete
Trust ≥3 / ≥5 / ≥7 / ≥9	+10% / +25% / +50% / +75%
Tier B / A / S preference	+15% / +30% / +50%
Recency 12m / 6m / 90d / 30d / 7d	+5% / +10% / +15% / +20% / +30%
Demographic targeting	Weighted by dimension, capped at +30%
Direct / named query	+25% to +100%
Verbatim	+25%

Corps build this live through the calculator. Every filter updates the total before the query is funded.

Tier Staking Thresholds for Corp-ETH Eligibility

Corpus tier	Stake required
C	10,000 LGCY
B	25,000 LGCY
A	50,000 LGCY
S	100,000 LGCY

Corp ETH requires enough LGCY in the relevant tier, quality at or above 60, and whatever buyer-selected filters the query imposed. Trust is not a universal hard gate. It is a priced buyer filter.

Two-Bucket Staking Model

Tier eligibility is calculated from two buckets that sum together: the earned bucket and the manually-staked bucket.

The earned bucket and manually-staked bucket are summed for corp-ETH eligibility. Manual deposits count immediately, but each deposit tranche must sit 48 hours before withdrawal.

Important: the 48-hour rule is a withdrawal lock only, not an eligibility delay. The deposited amount counts toward the threshold as soon as it lands.

ETH Distribution Among Panelists

On fresh panels, corp ETH is distributed only to the top-N answers by quality, where N is the number of panelists the corp paid for.

Top-N ETH share formula user_eth_share = (user_quality / sum_of_top_N_qualities) × contractor_pool

LGCY is still earned by everyone above the quality floor. ETH is what gets filtered by the top-N cutoff on fresh panels. Path A uses rotation, not top-N re-ranking at every query.

Privacy and Verbatim Modes

Legaicy is pseudonymous by default, not anonymous. Verbatim remains an explicit opt-in mode with a 20-respondent release floor and a 1.25x user-side multiplier paired to a +25% corp-side premium.

Wallet as Identity

The user's wallet is treated as a soulbound professional identity. EXP, trust, vesting progress, and future reputation are tied to it.

Use a fresh wallet when possible.
Prefer a hardware wallet.
Never approve unknown contracts.
Lost wallet access means lost EXP forever.

Hybrid Architecture

On-chain handles subscription, token emission, staking, trust, and payout surfaces. Off-chain handles app UX, Supabase storage, Haiku scoring, AI semantic search, and privacy-aware orchestration.

LGCY Token and Supply

LGCY has a 1,000,000,000 hard cap and a 5,000,000 pre-mint: 1,000,000 for LP bootstrap and 4,000,000 for the Protocol Vault reserve.

Initial LP target: 0.3 ETH + 1,000,000 LGCY at an implied starting price of $0.001 / LGCY.

Halvings at 100M, 250M, and 500M LGCY emitted. Past 500M, emissions follow an asymptotic decay curve. Total supply approaches 1B asymptotically; the 1B cap is enforced as a hard stop (post-cap, StakingVault.creditEarnings becomes a silent no-op for new mints; existing balances continue to vest, remain withdrawable, and count toward tier eligibility).

LGCY is disinflationary in the early years and becomes deflationary later. The halving curve and the ProtocolVault burn mechanic together produce a supply that grows slower over time, then shrinks once burns outpace emissions. Early years still emit more than they burn. Long term, the math flips.

Transparent supply tracking

Standard ERC-20 totalSupply() overstates "circulating" for protocols with vesting, staking, and reserves. LGCYToken exposes three distinct supply views, all anyone-callable:

totalSupply() — standard mintedAllTime − burnedAllTime. What aggregators read by default. ProtocolVault auto-mode mcap reads totalSupply() − balanceOf(DEAD).
circulatingSupply() — totalSupply() minus protocol-held / locked balances (ProtocolVault, StakingVault earned-locked, StakingVault manual-pending, SlashingManager burn queue). Useful for dashboards / aggregator listings that want to discount locked supply.
liquidSupply() — circulatingSupply() minus user-side stake balances still in withdrawal cooldown. The subset that could move to market right now.

Cumulative counters mintedAllTime() and burnedAllTime() are also exposed. All of these are Bucket 3 — locked forever once deployed. The contract interface is immutable; the views must ship at launch.

Revenue and Value Flow

User ETH follows the locked split of 15 / 85 in Years 0-2 and 5 / 95 in Year 3+. Corp ETH follows 15 / 80 / 2.5 / 2.5 in Years 0-2 and 5 / 80 / 7.5 / 7.5 in Year 3+. The specific 2.5% / 7.5% split values are Bucket 2 parameters — DAO-tunable within the Bucket 3 floors (contractor ≥80%, dev ≥5%). The floors are locked; the splits within them are tunable.

The contractor floor remains 80% and the dev/operator floor remains 5%.

Protocol Vault and Market Operations

The Protocol Vault remains the constrained execution surface for protocol-owned liquidity and token market operations. It is not an unrestricted operator wallet.

Four buttons, two categories

The vault exposes four anyone-callable button actions. Two are always callable. Two are mode-gated, with the contract enforcing that only one of them is active at a time.

Always callable (no mode gating): Claim LP Rewards; Burn LGCY.
Mode-gated (one active at a time): Add Liquidity (callable in LP MODE); Buy & Burn (callable in BURN MODE).

Auto-mode hysteresis

LP MODE (default): Add Liquidity callable, Buy & Burn reverts. Persists while LP value is below 3% of LGCY market cap. Transitions to BURN MODE when LP value crosses 3% mcap.

BURN MODE: Buy & Burn callable, Add Liquidity reverts. Persists while LP value is above 2% of LGCY market cap. Transitions to LP MODE when LP value crosses 2% mcap.

The 2% / 3% gap is hysteresis — it prevents rapid flip-flopping at a single threshold. The vault self-balances between 2–3% LP depth forever, with deflation happening during BURN MODE and LP deepening happening during LP MODE.

mcap is computed as (totalSupply() − balanceOf(DEAD)) × Uniswap V3 TWAP price. LGCYToken.totalSupply() is OZ-standard (NOT overridden) — it returns total ever minted, since _burn is never called. Burns are transfers to DEAD, so balanceOf(DEAD) accumulates the burn count and the subtraction yields totalMinted − totalBurnedAllTime, which is the operator-approved "total minted minus total burned" definition. Industry-standard pattern for DEAD-burn tokens.

The 2% / 3% threshold values are Bucket 2 — admin-adjustable in Y0–Y1, DAO-adjustable in Y2+, and lockable forever once equilibrium is found. The hysteresis-gating mechanic itself, the four-action split, and the rule that mcap reads totalSupply() − balanceOf(DEAD) are Bucket 3 — locked forever, immutable post-deploy.

Add Liquidity — two internal phases

Independent of the LP / BURN mode-gating above, Add Liquidity itself runs in one of two internal phases that govern how it sources the LGCY side of the LP add. Phase here is not the same thing as Mode; Mode controls which button is callable, Phase controls how Add Liquidity works once called in LP MODE.

Phase 1 (default at deploy): pairs incoming ETH directly with the 4,000,000 LGCY reserve. No market buy. The reserve depletes one-way as Phase 1 deploys it.
Phase 2 (auto-triggered once the reserve effectively empties): vault-LGCY-first, then market-buy if needed, then pair. The contract aims for half-and-half pairing between ETH and an equivalent ETH-value of LGCY. It draws from vaultLgcyBalance first (saving swap fees when the vault has accumulated LGCY from claimed LP fees or slashing remainders); only if the vault is short does it swap from the vault's ETH balance, capped by the pool-swap impact cap (swapSlippageBp). If the cap binds, the swap partial-fills and unspent ETH stays in the vault. The LP-add then runs nftManager.increaseLiquidity with amount0Min / amount1Min set per the LP-pairing slippage tolerance (slippageBp) to protect against pool-ratio drift between the swap and the LP-add.

The transition uses a dust threshold of 1 LGCY (1e18 wei). Phase 1 ends when the reserve drops below the threshold. The remaining dust stays in the vault permanently — burning it would cost more in gas than the dust is worth (≈ $0.001 at launch price). The threshold is a safety mechanism, not an economic parameter, and is Bucket 3 — immutable post-deploy. The transition emits PhaseOneCompleted(uint256 dustRemaining) exactly once for transparency.

In Phase 1, Add Liquidity never reverts when reserve runs short. If the reserve has less LGCY than needed for the ETH on hand, the contract pairs all available reserve LGCY with the proportional ETH at current ratio, leaves leftover ETH in the vault for the next call, and lets the dust threshold trigger Phase 2 entry. From there, Phase 2's vault-LGCY-first algorithm can deploy the leftover ETH by buying LGCY off-market within the swap-impact cap.

Operational locks

The four-button vault has additional locks that govern its day-to-day cadence:

Per-action cooldowns — Add Liquidity 24h, Buy & Burn 24h, Claim LP Rewards 24h, Burn LGCY 24h. Each tracked independently. Admin-adjustable in Y0–Y1, DAO-adjustable in Y2+, lockable later (Bucket 2). Bounds: 1 hour minimum, 7 days maximum (Bucket 3 — DAO cannot escape this band).
Pool-swap impact cap (swapSlippageBp) — 3% default, used by Buy & Burn and Phase 2 Add Liquidity to bound the price impact of a single ETH → LGCY swap. Bounds: 0.5% minimum / 5% maximum (Bucket 3). Cap value within the band is Bucket 2 — admin-adjustable Y0–Y1, DAO-adjustable Y2+, lockable later. (Supersedes the prior MAX_SLIPPAGE_BP = 300 Bucket 3 constant; the operator opted to make both slippage parameters admin-toggleable.)
LP-pairing slippage tolerance (slippageBp) — 1% default for Add Liquidity. Bounds: 1% minimum / 3% maximum (Bucket 3). Tolerance value within the band is Bucket 2. Distinct from the swap-impact cap above: this one bounds pool-ratio drift between the swap and the LP-add (used as amount0Min / amount1Min on nftManager.increaseLiquidity).
State pokes — every Add Liquidity and Buy & Burn call refreshes ratio, mode, and phase state at entry. Anyone calling the action effectively maintains the contract's state. No external keeper or cron job is needed.
No caller rebates in V1. None of the four button actions pay rebates to the address that calls them. LGCY holders benefit pro-rata from vault actions and have natural incentive to call; cooldowns ensure each call accumulates enough impact to justify the gas. Future contract upgrade may add rebates if real-world data shows actions going uncalled due to gas costs.
Claim LP Rewards routing. Claimed ETH fees stay in the vault's ETH balance, available as fuel for the next Add Liquidity or Buy & Burn cycle. Claimed LGCY fees route to vaultLgcyBalance, available for the next Burn LGCY action or for pairing in Add Liquidity. No fees flow to dev, treasury, or caller. Pure vault recycling.

Parameter Governance Model

Legaicy's parameters fall into three buckets that determine whether and how they can change after mainnet deploy. The split exists so trust-critical guarantees stay permanent while market-responsive numbers stay tunable.

Bucket 3 — locked forever

Permanent trust commitments. Cannot be changed by anyone, ever, after deploy.

1,000,000,000 LGCY hard cap.
5,000,000 LGCY pre-mint amount (encoded in LGCYToken constructor as a single mint to deployer; distribution to LP seed + ProtocolVault is operational per the deploy runbook, not contract-enforced).
Halving thresholds at 100M / 250M / 500M LGCY emitted; asymptotic curve past 500M.
Contractor share floor: ≥80% of corp ETH.
Dev / operator share floor: ≥5% of any ETH flow.
The structure of the seven-factor emission formula, the hybrid trust formula, and the two-bucket tier-staking model. The multipliers and counters exist; their values are Bucket 2.

Bucket 2 — adjustable now, permanently lockable later

Tunable during the protocol's early years while finding equilibrium. Power sits with admin in Y0–Y1, then transitions to DAO. Either admin (before relinquishing control) or DAO (via vote) can permanently lock any Bucket 2 parameter, after which it joins Bucket 3 forever.

Revenue splits, subject to Bucket 3 floors (currently 15/85 user Y0–2, 15/80/2.5/2.5 corp Y0–2).
Quality threshold (currently 60/100).
Trust progression numbers (currently 10/20/30/.../90 clean days + clean answers per level).
Verbatim multiplier (currently 1.25x).
Corpus tier multipliers (currently C = 1.0, B = 1.5, A = 2.0, S = 2.5).
demoMult coefficient and cap (currently 0.5 per filledPct, capped at 1.5x).
bonus_rate K constant (currently 10^12).
trustMult linear coefficient (currently 0.1 per level).

Bucket 1 — permanently flexible

These cannot be locked because real-world conditions keep moving — ETH price, LGCY market price, market dynamics. Locking them would invalidate the protocol's ability to function across changing conditions. Power sits with admin in Y0–Y1, then transitions to DAO permanently. There is no lock path.

Subscription price (currently 0.005 ETH/year Y1 promo, 0.01 ETH/year standard).
Question fee starting price (currently 0.0001 ETH for Q1 of the daily cycle, doubling per question).
Tier staking thresholds (currently 10K / 25K / 50K / 100K LGCY for C / B / A / S).

Lock rules

Bucket 3 cannot be unlocked by anyone, ever. The Bucket 2 → Bucket 3 transition is one-way: once locked, it stays locked. Bucket 1 cannot be locked even by DAO vote. Bucket 3 floors constrain Bucket 2 changes — no DAO vote can move contractor share below 80% or dev share below 5%. Contracts enforce these floors as hard require() checks even on admin-controlled adjustment functions.

Voting weight

DAO governance weight in Y2+ derives from EXP, not from LGCY token balance held.

EXP is the receipt of every clean answer ever submitted, accumulated 1:1 with lifetime LGCY earned. It is non-transferable and soulbound to the earning wallet — it cannot be bought, sold, or merged across wallets, and it is lost permanently if the user loses access to their wallet.

Holding LGCY — whether purchased on the open market, transferred from a contributor, or vested out of the staking vault — grants zero voting power on its own. The only way to gain governance weight is to do clean work. Governance is unbuyable: whales cannot purchase voting power and speculators cannot direct protocol evolution.

The specific weighting curve (linear EXP, sqrt(EXP), tiered, or hybrid) is parked for the future DAO design session. The principle that vote weight derives from EXP rather than LGCY balance is itself a Bucket 3 commitment; the curve formula is Bucket 2 — adjustable until locked.

Centralization at Launch

Year 0 is still openly centralized across trust writing, slashing and review, payout publication, admin roles, and LP custody. The credibility test is not whether that reality exists, but whether it is disclosed and then reduced over time.

Progressive Decentralization

Year 0 prioritizes launch safety. Year 1 adds licensing and hardening. Year 2 shifts more responsibility outward. Year 3+ activates the locked lower operator share and higher protocol / treasury split.

Open Ideas and Parked Enhancements

Some corp-side enhancements are explicitly parked, not locked. The current examples are:

corp burn-for-discount mechanics using a quote-and-lock flow;
corp XP / loyalty rewards;
stronger on-chain soulbound EXP enforcement.

These are future possibilities, not present product commitments.

Legal and Policy Posture

Legaicy is framed as compensated digital labor. It avoids passive-yield language and ties earning to work, quality, consent, and visible settlement rules.

Risks and Open Execution Work

Product risks

Corp demand may develop slower than the corpus supply side.
Onboarding may still be too complex for some users.
Verbatim and privacy mistakes would be high-cost failures.

System risks

Bot pressure remains constant.
Token incentive tuning may require repeated calibration.
Year 0 centralization is still a real trust dependency.

Current Status

Subscription is live on Sepolia. The wider contract suite is designed but not fully deployed. The app UI is in a strong mock state, and the data and contract plumbing are still pre-launch.

Mission

Legaicy's mission is to build durable infrastructure for paid human thought.

If AI makes content abundant, then trustworthy human judgment becomes scarce. Legaicy is designed to price, route, score, and settle that scarcity without pretending the system is simpler than it is.