# DeFiMind MCP server

Live LP analytics — Uniswap V2/V3, Balancer, Curve stableswap: PnL, health, slippage, depeg risk.

## Links
- Registry page: https://www.getdrio.com/mcp/io-github-defimind-ai-defimind-mcp
- Repository: https://github.com/defimind-ai/defimind-mcp
- Website: https://defimind.ai

## Install
- Endpoint: https://mcp.defimind.ai/mcp
- Auth: Not captured

## Setup notes
- Remote endpoint: https://mcp.defimind.ai/mcp

## Tools
- AnalyzePosition - Analyze why a Uniswap V2 or V3 LP position is gaining or losing money. Decomposes PnL into impermanent loss, accumulated fees, and net result, with optional real APR if a holding period is supplied. Returns current value, hold value, IL percentage, fee income, net PnL, real APR, and a diagnosis label. Endpoint: https://mcp.defimind.ai/mcp
- AnalyzeBalancerLP - Analyze a 2-asset Balancer weighted-pool LP position's PnL. Decomposes impermanent loss using the weighted-pool formula where the base token's weight affects IL magnitude. Values are denominated in opp-token units per BalancerImpLoss's convention; fee income is not attributed in v1 because Balancer pools only expose vault-level fees with no per-LP attribution. Endpoint: https://mcp.defimind.ai/mcp
- AnalyzeStableswapLP - Analyze a 2-asset Curve-style Stableswap LP position's PnL using the amplified-invariant IL formula where small depegs can produce surprisingly large IL at high A. Values are in peg-numeraire (tokens valued 1:1); fee income is not attributed in v1 (pool-global only); positions whose implied alpha is in the unreachable regime return None on il_percentage, net_pnl, and real_apr. Endpoint: https://mcp.defimind.ai/mcp
- SimulatePriceMove - Project a Uniswap V2 or V3 LP position's value at a hypothetical price change from the CURRENT pool state (not from entry). A price_change_pct of -0.30 asks 'what if price drops 30% from here'. Returns new value, IL at the simulated price, and percentage change in position value. Fee projection is not modeled (always null). Endpoint: https://mcp.defimind.ai/mcp
- SimulateBalancerMove - Project a 2-asset Balancer weighted-pool LP position's value at a hypothetical price change from the CURRENT pool state. The shock is applied to the base-token price in opp units; IL depends on both the shock magnitude and the pool's weights. Returns new value in opp-numeraire, IL at the simulated price, and percentage change. Endpoint: https://mcp.defimind.ai/mcp
- SimulateStableswapMove - Project a 2-asset Curve-style Stableswap LP position's value at a hypothetical depeg from the CURRENT pool state. The shock multiplies the pool's current alpha by (1 + price_change_pct); at high A, large shocks may be physically unreachable and in that case new_value, il_at_new_price, and value_change_pct are returned as null. Values are in peg-numeraire. Endpoint: https://mcp.defimind.ai/mcp
- CheckPoolHealth - Snapshot pool-level health metrics for a Uniswap V2 or V3 pool: TVL in token0 numeraire, reserves, accumulated fees, LP concentration, and swap activity. Answers 'is this a pool I would deposit into?' at the pool level (not position level). num_swaps and fee_accrual_rate_recent are V2-only; V3 returns null for these because V3 has no per-swap history array. A single-block live snapshot also returns null for the LP-concentration and swap metrics (unrecoverable from state alone). Endpoint: https://mcp.defimind.ai/mcp
- DetectRugSignals - Detect rug-pull signals on a Uniswap V2 or V3 pool via three threshold checks: suspiciously low TVL, top-LP concentration above a limit, and inactive-pool-with-liquidity. Composes over CheckPoolHealth and returns per-signal booleans plus a count-based risk level (low/medium/high/critical). The inactive-with-liquidity signal is V2-only; V3 pools report False for it with a note in details. Endpoint: https://mcp.defimind.ai/mcp
- CalculateSlippage - Calculate slippage and price-impact decomposition for a proposed swap on a Uniswap V2 or V3 pool. Returns spot vs execution price, slippage percentage, slippage cost in output-token units, and price impact. Also returns the maximum trade size that stays within 1% slippage for V2 pools; V3 returns null for that field because tick-crossing math has not yet been inverted. Endpoint: https://mcp.defimind.ai/mcp
- AssessDepegRisk - Quantify a 2-asset Curve-style Stableswap LP position's exposure to a stablecoin depeg. Computes IL at multiple depeg levels (default 2%, 5%, 10%, 20%, 50%) via the closed-form stableswap-invariant expansion, with an optional V2 constant-product benchmark at each level. Some depeg levels are physically unreachable at high A — unreachable scenarios return null on il_pct, lp_value_at_depeg, and hold_value_at_depeg; the V2 benchmark stays populated. Endpoint: https://mcp.defimind.ai/mcp
- BuildStateTwin - Read a pool's on-chain state once and return it as a serialized State Twin (JSON): the protocol-specific snapshot plus a content_hash, in the wire form a client rehydrates locally. Use this to pull a single managed-RPC twin, then run any number of counterfactuals (price moves, IL, slippage) client-side, off this server. Covers Uniswap V2/V3, Balancer 2-asset weighted, and Curve 2-asset plain stableswap pools. The endpoint stores and logs nothing — your rpc_url is never persisted. Endpoint: https://mcp.defimind.ai/mcp

## Resources
Not captured

## Prompts
Not captured

## Metadata
- Owner: io.github.defimind-ai
- Version: 0.2.3
- Runtime: Streamable Http
- Transports: HTTP
- License: Not captured
- Language: Not captured
- Stars: Not captured
- Updated: Jun 22, 2026
- Source: https://registry.modelcontextprotocol.io