> For the complete documentation index, see [llms.txt](https://onex-3.gitbook.io/onex-document/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://onex-3.gitbook.io/onex-document/coin/pillars-12-14.md).

# Pillars 12 14

## Pillar 12: Community Loyalty Reward Amplifier (CLRA)

### Loyalty Scoring System (0–1,000 points)

**Earning Loyalty Points:**

| Action                           |                Points Earned |
| -------------------------------- | ---------------------------: |
| Holding 1,000 ONEx for 1 day     |                        +1 LS |
| Staking 1,000 ONEx for 1 day     |                        +3 LS |
| Providing Liquidity for 1 day    |         +5 LS per 1,000 ONEx |
| Holding in IEV (not exiting)     |         +2 LS per 1,000 ONEx |
| Buying ONEx                      |       +10 LS per $100 bought |
| Participating in governance vote |              +20 LS per vote |
| Referring new holders            | +50 LS per verified referral |

**Losing Loyalty Points:**

| Action                         |          Points Lost |
| ------------------------------ | -------------------: |
| Selling ONEx                   | -50 LS per $100 sold |
| Exiting IEV early              |              -100 LS |
| Unstaking early                |               -30 LS |
| Removing Liquidity             |               -40 LS |
| Triggering whale restriction   |              -200 LS |
| Selling during circuit breaker |  -500 LS (immediate) |

### Loyalty Tiers & Benefits

| Score    | Tier        | APY Multiplier | Sell Tax Discount | Other Benefits                                           |
| -------- | ----------- | -------------: | ----------------: | -------------------------------------------------------- |
| 0–100    | 🥉 Bronze   |           1.0× |              None | Base staking APY only                                    |
| 101–300  | 🥈 Silver   |           1.2× |               10% | Priority support                                         |
| 301–500  | 🥇 Gold     |           1.5× |               25% | Early feature access                                     |
| 501–750  | 💎 Platinum |           2.0× |               40% | Governance vote boost                                    |
| 751–1000 | 👑 Diamond  |           3.0× |               60% | Priority IEV queue + governance boost + exclusive access |

### Staking Tiers (Base APY)

| Lock Period                | Base APY |
| -------------------------- | -------: |
| Flexible (unstake anytime) |       8% |
| 30-Day Lock                |      15% |
| 90-Day Lock                |      22% |
| 180-Day Lock               |      30% |
| 365-Day Lock               |      40% |

**Diamond holder on 365-Day Lock: 40% × 3.0× = 120% APY**

All rewards vest linearly over 7 days after claim. Selling before fully vested forfeits unvested rewards — returned to staking pool.

***

## Pillar 13: ZK Privacy Shield (ZKPS) ★ NEW

*Entirely new — not present in v1.0. ZK-proof technology applied to order privacy.*

### Core Concept

ZKPS uses zero-knowledge proofs to **hide the size and intent of large orders** from MEV bots, front-runners, and market manipulators. Large sell orders are split into encrypted fragments and executed across multiple blocks without revealing the full position to the mempool.

### How ZKPS Works

1. User submits large sell order to ZKPS contract (not directly to DEX)
2. ZKPS generates ZK proof that: (a) user holds the claimed tokens, (b) order is valid, (c) taxes are pre-computed correctly
3. Order is split into random-sized fragments (between 500–5,000 ONEx each)
4. Fragments execute across random block intervals (1–5 blocks apart)
5. Each fragment appears as a normal retail-sized sell — indistinguishable from organic activity
6. ZK proof submitted on-chain — verifiably correct without revealing full order size

### ZKPS Benefits

| Benefit                     | Description                                           |
| --------------------------- | ----------------------------------------------------- |
| Front-running impossible    | MEV bots cannot see large order in mempool            |
| Sandwich attacks impossible | Random fragment sizes prevent profiling               |
| Price impact reduced        | Small fragments have negligible individual impact     |
| Fully verifiable            | ZK proof ensures no double-spending or tax evasion    |
| Privacy-preserving          | Only the sender and ZKPS contract know the full order |

### ZKPS Technical Stack

| Component          | Specification                                    |
| ------------------ | ------------------------------------------------ |
| Proof system       | Groth16 (SNARK) — low verification gas cost      |
| Circuit            | Written in Circom, audited independently         |
| Verifier contract  | Deployed on ONEx Chain (\~200,000 gas per proof) |
| Maximum order size | 1,000,000 ONEx (beyond this, SEM required)       |

***

## Pillar 14: AI Anomaly Detection Engine (AADE) ★ NEW

*Entirely new. Machine learning-based real-time attack detection and automated response.*

### Core Concept

AADE runs a continuously-trained anomaly detection model that **identifies attack patterns before they execute**, enabling pre-emptive defense rather than reactive defense.

### AADE Detection Capabilities

| Attack Type                   | Detection Method                                                                   |
| ----------------------------- | ---------------------------------------------------------------------------------- |
| Coordinated dump (5+ wallets) | Graph neural network — detects correlated sell timing within 60s                   |
| Wash trading                  | Circular transaction detection — identifies buy/sell loops between related wallets |
| Pump orchestration            | Unusual buy clustering before suspected dump — flagged 24–48 hours ahead           |
| Sybil governance attack       | Voting pattern analysis — correlates new wallet creation with vote timing          |
| Flash loan prep               | Large borrow events on lending protocols → pre-alerts OBCB oracle system           |
| Bridge exploit attempt        | Anomalous bridge transaction patterns — flags to Bridge Guardian multisig          |
| Bot swarm attack              | Gas price clustering and timing regularity → identifies automated actors           |

### AADE Response Actions

| Confidence Level       | Response                                                                            |
| ---------------------- | ----------------------------------------------------------------------------------- |
| Low confidence anomaly | Increase DSPA monitoring frequency to every block                                   |
| Medium confidence      | Raise WDTS tier for flagged wallets by one level; increase sell tax by +2%          |
| High confidence        | Freeze flagged wallets from selling for 1 hour; alert governance; deploy ABR Mode 2 |
| Critical confidence    | Activate OBCB Level 1 pre-emptively; flag wallets for full WDTS investigation       |

### AADE Model Architecture

| Component           | Specification                                                                                                     |
| ------------------- | ----------------------------------------------------------------------------------------------------------------- |
| Model               | Isolation Forest + GNN (Graph Neural Network) ensemble                                                            |
| Training data       | On-chain transaction history, updated every 6 hours                                                               |
| Inference           | Off-chain on decentralized compute (Gensyn or similar) — result posted on-chain as signed attestation             |
| Challenge mechanism | Any ONEx staker can challenge an AADE decision within 4 hours; successful challenge rewards challenger 1,000 ONEx |
| Bias protection     | Model cannot be trained by team unilaterally — training jobs posted publicly with 24-hour review window           |