AI Endpoints

QsnDEX integrates AI-powered analytics across multiple endpoints. These features leverage Groq-hosted LLMs (Llama 3.3 70B, Llama 3.1 8B, DeepSeek R1 70B) to provide real-time insights on tokens, pools, and trading conditions.

AI Chat

POST /api/ai/chat

Multi-turn conversational AI assistant with context about DeFi, QsnDEX features, and on-chain data.

Request Body:

{
  "messages": [
    { "role": "user", "content": "What is impermanent loss?" },
    { "role": "assistant", "content": "Impermanent loss occurs when..." },
    { "role": "user", "content": "How can I minimize it?" }
  ],
  "language": "en"
}

Response:

{
  "success": true,
  "data": {
    "response": "To minimize impermanent loss, you can...",
    "model": "llama-3.3-70b-versatile"
  }
}

Token Analysis

POST /api/ai/analyze/token

Deep AI-driven analysis of a token, including on-chain metrics, risk factors, and market outlook.

Request Body:

{
  "token_address": "0x...",
  "chain_id": 167000
}

Pool Analysis

POST /api/ai/analyze/pool

AI analysis of a liquidity pool with yield projections, risk assessment, and optimization suggestions.

Request Body:

{
  "pool_address": "0x...",
  "chain_id": 167000
}

Swap Insight

POST /api/ai/swap-insight

Sentiment and market analysis for a specific swap pair.

Request Body:

{
  "token_in": "0x...",
  "token_out": "0x...",
  "chain_id": 167000
}

Anti-Rug Shield

GET /api/safety/{token}

Performs 8 on-chain safety checks on a token and returns a composite safety score.

Path Parameters:

Parameter	Type	Description
token	string	Token contract address

Query Parameters:

Parameter	Type	Description
chain_id	integer	Target chain ID

Response:

{
  "success": true,
  "data": {
    "token": "0x...",
    "score": 8,
    "max_score": 10,
    "checks": {
      "ownership_renounced": true,
      "mint_disabled": true,
      "liquidity_locked": true,
      "honeypot_check": false,
      "proxy_contract": false,
      "high_tax": false,
      "verified_source": true,
      "sufficient_liquidity": true
    },
    "risk_level": "LOW"
  }
}

Safety Checks

Check	Description
ownership_renounced	Contract ownership has been renounced
mint_disabled	No minting function or minting is disabled
liquidity_locked	LP tokens are locked in a timelock contract
honeypot_check	Token can be freely bought and sold
proxy_contract	Whether the token uses an upgradeable proxy
high_tax	Transfer tax exceeds acceptable threshold
verified_source	Source code is verified on the block explorer
sufficient_liquidity	Pool has adequate liquidity depth

Smart Entry Signal

GET /api/entry-signal/{token}

Technical analysis signal for a token, combining RSI, momentum, and trend indicators.

Path Parameters:

Parameter	Type	Description
token	string	Token contract address

Query Parameters:

Parameter	Type	Description
chain_id	integer	Target chain ID

Response:

{
  "success": true,
  "data": {
    "token": "0x...",
    "signal": "BUY",
    "confidence": 0.78,
    "rsi": 32.5,
    "momentum": "bullish",
    "trend": "upward",
    "summary": "RSI indicates oversold conditions with improving momentum."
  }
}

Autopilot

GET /api/autopilot/{wallet}

Analyzes a wallet's portfolio health and generates actionable suggestions for optimization.

Path Parameters:

Parameter	Type	Description
wallet	string	User wallet address

Query Parameters:

Parameter	Type	Description
chain_id	integer	Target chain ID

Response:

{
  "success": true,
  "data": {
    "wallet": "0x...",
    "health_score": 72,
    "risk_level": "MODERATE",
    "suggestions": [
      "Consider diversifying: 85% of portfolio is in a single asset.",
      "WETH/USDC pool has higher APR than your current LP position.",
      "Token XYZ has a low safety score (3/10). Review your holding."
    ]
  }
}

Rate Limits

All AI endpoints are rate-limited per wallet or IP address using Redis:

Endpoint	Rate Limit
POST /api/ai/chat	10 requests per hour
POST /api/ai/analyze/token	10 requests per hour
POST /api/ai/analyze/pool	10 requests per hour
POST /api/ai/swap-insight	10 requests per hour
GET /api/safety/{token}	20 requests per hour
GET /api/entry-signal/{token}	20 requests per hour
GET /api/autopilot/{wallet}	5 requests per hour

Exceeding the limit returns HTTP 429 with a Retry-After header indicating when the next request can be made.