# HyperSignal MCP (remote · hypersmash.fly.dev)

Hyperliquid analytics for AI agents: whale flow, liquidations, funding, risk, signed signals.

- Trust score: 67/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `hypersmash.fly.dev`: 67/100 (this document), [markdown](https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersmash.md), [page](https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersmash)
- npm · `hypersignal-mcp`: 33/100, [markdown](https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersignal-mcp.md), [page](https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersignal-mcp)

## Channel facts

- Endpoint: `https://hypersmash.fly.dev/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.14.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 57/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (hl_delete_alert).
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 78/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 4408 tokens (~129/item across 34 items; 30 tools + 4 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 23/100
  - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 90/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 66% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 3 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 32 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the HyperSignal MCP server?

HyperSignal MCP is a hosted endpoint at https://hypersmash.fly.dev/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http johnsmithxy1mmm-sys-hypersignal-mcp 'https://hypersmash.fly.dev/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "johnsmithxy1mmm-sys-hypersignal-mcp": {
      "url": "https://hypersmash.fly.dev/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "johnsmithxy1mmm-sys-hypersignal-mcp": {
      "type": "http",
      "url": "https://hypersmash.fly.dev/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.johnsmithxy1mmm-sys-hypersignal-mcp]
url = "https://hypersmash.fly.dev/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "johnsmithxy1mmm-sys-hypersignal-mcp": {
      "type": "remote",
      "url": "https://hypersmash.fly.dev/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add johnsmithxy1mmm-sys-hypersignal-mcp --url 'https://hypersmash.fly.dev/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  johnsmithxy1mmm-sys-hypersignal-mcp:
    url: "https://hypersmash.fly.dev/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "johnsmithxy1mmm-sys-hypersignal-mcp": {
      "Transport": "http",
      "Url": "https://hypersmash.fly.dev/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add johnsmithxy1mmm-sys-hypersignal-mcp -t streamable-http -u 'https://hypersmash.fly.dev/mcp'
```

### Other

```json
{
  "mcpServers": {
    "johnsmithxy1mmm-sys-hypersignal-mcp": {
      "type": "http",
      "url": "https://hypersmash.fly.dev/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-02 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 10 to 13. That category is still filling its 30-day observation window: 3 days of observed history at the previous scan, 4 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-31 (score 65, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-30 (score 64, 0)

- [functional improvement] Stability: unverified → 0.03
- [functional] Server version: 1.14.0 → 1.16.0
- [functional] New tool “hl_buy_pro”

### 2026-10-29 (score 64)

First indexed and scored.

## MCP tools (30)

### `hl_request_free_key` (~124 tokens)

Get a free API key

Issue yourself a free API key, instantly and with no signup. The key grants 100 premium calls per month — enough to evaluate every premium tool. Send it as the X-API-Key header on later requests. Call this first if a premium tool says it needs a key. One active key per requester: calling again replaces the previous key and keeps the same monthly usage, so it is safe to call if you lost your key but is not a way to reset the quota.

Input parameters:

- `acknowledge` (boolean): Acknowledges that analytics are informational and not investment advice.

Output parameters:

- `apiKey` (string)
- `header` (string)
- `monthlyPremiumCalls` (number)
- `reissued` (boolean)
- `tier` (string)
- `upgrade` (string)
- `usedThisMonth` (number)

### `hl_buy_pro` (~153 tokens)

Buy Pro (x402)

Buy 30 days of Pro (unlimited premium calls) for $29 in USDC, paid over x402 on Base. Without an X-API-Key it issues a new Pro key; with one it renews that key (a free key is upgraded in place), adding 30 days to its current end. The first call answers with the payment requirements; pay and retry with the X-PAYMENT header. The payment is taken only once the key is ready. No automatic refunds. If the answer says the payment could not be confirmed, do not pay again: quote its reference. How to pay: https://github.com/johnsmithxy1mmm-sys/HyperliquidMCP#pricing

Output parameters:

- `apiKey` (string)
- `daysAdded` (number)
- `expiresAt` (string)
- `header` (string)
- `reference` (string)
- `renewed` (boolean)
- `tier` (string)

### `hl_get_markets` (~148 tokens)

List Hyperliquid perp markets

List Hyperliquid perpetual markets with mark/oracle price, hourly + annualized funding, open interest, and 24h volume. Use to discover valid coin symbols and scan the board. Supports name filter, sorting, and pagination.

Input parameters:

- `dex` (string): HIP-3 builder dex to list (e.g. 'xyz'); omit for the main dex.
- `filter` (string): Case-insensitive substring match on coin symbol (e.g. 'BTC').
- `includeDelisted` (boolean): Include delisted markets.
- `limit` (integer)
- `offset` (integer)
- `sort` (string): Sort field (desc for numeric, asc for name).

Output parameters:

- `limit` (number)
- `markets` (array)
- `nextOffset` (number|null)
- `offset` (number)
- `total` (number)

### `hl_get_orderbook` (~98 tokens)

Get L2 order book

Level-2 order book for a coin: top-N bid/ask levels with price, size, and order count, plus spread and mid. Use for microstructure, spread, and near-touch liquidity. If the coin is unknown, call hl_get_markets first.

Input parameters:

- `coin` (string, required): Perp coin symbol, e.g. 'BTC' (case-insensitive).
- `depth` (integer): Number of levels per side.

Output parameters:

- `asks` (array)
- `bids` (array)
- `coin` (string)
- `midPx` (number|null)
- `spread` (number|null)
- `spreadBps` (number|null)
- `time` (number)

### `hl_get_candles` (~150 tokens)

Get OHLCV candles

OHLCV candlestick history for a coin at a given interval. Provide either an explicit [startTime,endTime] (ms epoch) or a `lookback` count of the most recent candles. Returns most-recent-last, paginated.

Input parameters:

- `coin` (string, required): Perp coin symbol, e.g. 'ETH'.
- `endTime` (integer): End time in ms epoch (default now).
- `interval` (string)
- `limit` (integer)
- `lookback` (integer): Number of most-recent candles to fetch (ignored if startTime given).
- `offset` (integer)
- `startTime` (integer): Start time in ms epoch.

Output parameters:

- `candles` (array)
- `coin` (string)
- `interval` (string)
- `nextOffset` (number|null)
- `total` (number)

### `hl_get_funding_history` (~117 tokens)

Get funding rate history

Historical hourly funding rates for a coin over a time window, with cumulative and average annualized funding. Use for carry analysis and funding trend. Defaults to the last 7 days if no window is given.

Input parameters:

- `coin` (string, required): Perp coin symbol, e.g. 'BTC'.
- `endTime` (integer): End time ms epoch (default now).
- `limit` (integer)
- `offset` (integer)
- `startTime` (integer): Start time ms epoch (default: 7 days ago).

Output parameters:

- `avgApr` (number)
- `avgHourly` (number)
- `coin` (string)
- `count` (number)
- `cumulative` (number)
- `entries` (array)
- `nextOffset` (number|null)

### `hl_get_account` (~109 tokens)

Get account positions & margin

Perp account snapshot for any address: equity, margin usage, withdrawable, and open positions (size, entry, uPnL, leverage, liquidation price). Read-only; works for any public 0x address.

Input parameters:

- `address` (string, required): EVM 0x address (42 hex chars).
- `dex` (string): Perp dex: omit for the main dex, a HIP-3 builder dex name (e.g. 'xyz'), or 'all' for every dex.

Output parameters:

- `accountValue` (number)
- `address` (string)
- `marginUtilization` (number)
- `positions` (array)
- `totalMarginUsed` (number)
- `totalNtlPos` (number)
- `withdrawable` (number)

### `hl_get_open_orders` (~129 tokens)

Get open orders

Open resting orders for an address: coin, side, price, size, order id, type, and trigger info. Read-only; paginated. Use to inspect any wallet's live order book footprint.

Input parameters:

- `address` (string, required): EVM 0x address.
- `coin` (string): Optional filter to a single coin.
- `dex` (string): Perp dex: omit for the main dex, a HIP-3 builder dex name (e.g. 'xyz'), or 'all' for every dex.
- `limit` (integer)
- `offset` (integer)

Output parameters:

- `address` (string)
- `nextOffset` (number|null)
- `orders` (array)
- `total` (number)

### `hl_signal_pubkey` (~68 tokens)

Get signal verification public key

Returns the server's Ed25519 public key (SPKI DER, base64) used to sign emitted signals, so anyone can independently verify a signal's authenticity and timestamp. Canonicalization: JSON with recursively sorted keys over {payload, ts}; signature is base64 Ed25519.

Output parameters:

- `alg` (string)
- `canonicalization` (string)
- `ephemeral` (boolean)
- `publicKey` (string)

### `hl_link_telegram` (~58 tokens)

Link Telegram for alerts

Get a one-time Telegram link (valid 15 minutes). Open it and press Start: this key's standing alerts are then also pushed to that chat (up to 30 messages an hour; everything stays available via hl_poll_alerts).

Output parameters:

- `expiresInMinutes` (number)
- `link` (string)

### `hl_unlink_telegram` (~29 tokens)

Unlink Telegram

Stop pushing this key's alerts to Telegram (they remain available via hl_poll_alerts).

Output parameters:

- `unlinked` (boolean)

### `hl_set_webhook` (~79 tokens)

Set alert webhook

POST this key's fired alerts to an https URL (public addresses only), up to 30 an hour. Each body is signed: header x-hypersignal-signature = sha256=HMAC-SHA256(secret, body). The secret is returned only once.

Input parameters:

- `url` (string, required): https URL that receives alert POSTs.

Output parameters:

- `secret` (string)
- `url` (string)

### `hl_delete_webhook` (~30 tokens)

Delete alert webhook

Stop POSTing this key's alerts to its webhook (they remain available via hl_poll_alerts).

Output parameters:

- `deleted` (boolean)

### `hl_score_calibration` (~135 tokens)

Smart-money score calibration

Out-of-sample evidence for whether the hl_smart_money_score actually predicts anything: Spearman rank correlation between each wallet's score and the realized return it went on to earn over the forward window (PnL over account value), plus mean forward return per score quartile. Verdicts: insufficient_data (not enough resolved observations yet), inverted (significantly NEGATIVE — higher-scored wallets did worse), no_evidence, weak_positive, positive (both require p<0.05). Only server-selected cohorts enter the sample, so a caller cannot steer this figure. Read this before treating the score as a forecast.

Output parameters:

- `methodology` (string)
- `n` (number)
- `observations` (object)
- `pValue` (number|null)
- `quartiles` (array)
- `spearman` (number|null)
- `verdict` (string)

### `hl_whale_positions` (~219 tokens)

Aggregate whale positions by coin

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Aggregated positioning of a whale cohort for a coin (or all coins): total long/short size & notional, long/short wallet counts, weighted-average entry, net bias, and change vs 1h/24h ago. Cohort = `cohort` arg or HL_WHALE_ADDRESSES. Analytics only, not investment advice.

Input parameters:

- `cohort` (array): Explicit list of 0x wallet addresses.
- `coin` (string): Restrict to one coin (a HIP-3 coin like 'xyz:XYZ100' reads its own dex); omit for all coins.
- `dex` (string): Perp dex: omit for the main dex, a HIP-3 builder dex name (e.g. 'xyz'), or 'all' for every dex (at most 20 wallets).
- `topN` (integer): Max wallets to include from the cohort.

Output parameters:

- `asOf` (number)
- `coins` (array)
- `dex` (string)
- `disclaimer` (string)
- `source` (string)
- `walletsCappedTo` (number|null)
- `walletsQueried` (number)
- `walletsRequested` (number)

### `hl_whale_flow_alerts` (~206 tokens)

Whale flow alerts (recent large moves)

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Recent large changes in whale cohort positioning: which coins saw net long/short notional shift beyond a USD threshold within a lookback window — 'who opened/closed what in the last hour'. Detects full exits. Seeds a baseline on first call. Analytics only, not investment advice.

Input parameters:

- `cohort` (array): Explicit 0x wallet addresses; else HL_WHALE_ADDRESSES.
- `dex` (string): Perp dex: omit for the main dex, a HIP-3 builder dex name (e.g. 'xyz'), or 'all' for every dex (at most 20 wallets).
- `lookbackMinutes` (integer): Compare against a snapshot ~this old.
- `thresholdUsd` (number): Min absolute net-notional change to alert on.
- `topN` (integer)

Output parameters:

- `alerts` (array)
- `baselineFound` (boolean)
- `dex` (string)
- `disclaimer` (string)
- `lookbackMinutes` (number)
- `source` (string)
- `walletsCappedTo` (number|null)
- `walletsQueried` (number)
- `walletsRequested` (number)

### `hl_liquidation_map` (~155 tokens)

Liquidation cluster map

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Estimated liquidation clusters for a coin from a whale cohort's open positions: distribution of liquidation prices and notional-at-risk bucketed by % distance from mark, plus nearest downside/upside cascade levels. Estimate over the cohort only (not exchange-wide). Analytics only, not investment advice.

Input parameters:

- `bucketPct` (number): Bucket width as % of mark price.
- `cohort` (array)
- `coin` (string, required): Perp coin symbol.
- `maxDistancePct` (number): Ignore liq prices beyond this % from mark.
- `topN` (integer)

Output parameters:

- `buckets` (array)
- `coin` (string)
- `disclaimer` (string)
- `markPx` (number)
- `nearestDownsideCluster`
- `nearestUpsideCluster`
- `positionsConsidered` (number)
- `totalNotionalAtRiskUsd` (number)
- `walletsQueried` (number)

### `hl_funding_screener` (~195 tokens)

Funding rate screener

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Screen every Hyperliquid perp by annualized funding: top markets by absolute funding (carry candidates), sign (longs-pay vs shorts-pay), and — when predictedFundings is available — cross-venue funding spreads for delta-neutral strategies. Filter by min |APR| and OI/volume. Analytics only, not investment advice.

Input parameters:

- `includeVenueSpreads` (boolean): Include cross-venue predicted funding spreads if available.
- `limit` (integer)
- `minAbsApr` (number): Minimum |annualized funding| as a fraction (0.05 = 5% APR).
- `minOpenInterestUsd` (number): Filter out thin markets below this OI (USD).
- `offset` (integer)
- `side` (string)

Output parameters:

- `disclaimer` (string)
- `matched` (number)
- `nextOffset` (number|null)
- `results` (array)
- `scannedMarkets` (number)
- `venueSpreads` (array)

### `hl_hip3_analytics` (~168 tokens)

HIP-3 market analytics

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Analytics for HIP-3 (builder-deployed) perp markets such as RWA/equity perps: volume, open interest, funding, and 24h change, enumerated across builder dexs. Optionally pass `spotRefs` (coin→external spot price) to compute basis vs an external reference. Analytics only, not investment advice.

Input parameters:

- `dex` (string): Filter to one builder dex by name (the part before ':' in market names).
- `limit` (integer)
- `offset` (integer)
- `spotRefs` (object): Optional map of coin symbol -> external spot price for basis calculation.

Output parameters:

- `dexsFailed` (array)
- `dexsNotScanned` (array)
- `dexsScanned` (array)
- `disclaimer` (string)
- `hip3MarketCount` (number)
- `markets` (array)
- `nextOffset` (number|null)

### `hl_portfolio_risk` (~167 tokens)

Portfolio risk report

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Risk report for a perp address: liquidation distance per position, effective leverage, concentration (HHI + largest position share), uniform stress scenarios (-5/-10/-20% and mirror), and Sharpe of realized PnL from recent fills. Analytics only, not investment advice.

Input parameters:

- `address` (string, required): EVM 0x address.
- `dex` (string): Perp dex: omit for the main dex, a HIP-3 builder dex name (e.g. 'xyz'), or 'all' for every dex.
- `shocks` (array): Price-shock fractions to stress test (applied uniformly across positions).

Output parameters:

- `accountValue` (number)
- `address` (string)
- `concentrationHhi` (number)
- `disclaimer` (string)
- `effectiveLeverage` (number)
- `largestPositionShare` (number)
- `marginUtilization` (number)
- `positions` (array)
- `realizedPnlSharpe` (number|null)
- `stressScenarios` (array)

### `hl_vault_screener` (~153 tokens)

Vault screener

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Screen Hyperliquid user vaults by APR, max drawdown, Sharpe/Sortino, age, TVL, and leader capital share. Provide `vaults` (list of vault 0x addresses) or set HL_VAULT_ADDRESSES. Metrics derived from vault history. Analytics only, not investment advice.

Input parameters:

- `minTvlUsd` (number)
- `sort` (string)
- `vaults` (array): Vault 0x addresses to screen (max 50); else HL_VAULT_ADDRESSES.

Output parameters:

- `disclaimer` (string)
- `screened` (number)
- `unavailable` (array)
- `vaults` (array)

### `hl_trader_report` (~183 tokens)

Trader report

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Behavioral report for any wallet from its fills: winrate, average R (avg win / avg loss), total realized PnL, PnL curve, holding-time style (scalp vs swing), fees paid, and most-traded coins. Lookback in days (Hyperliquid serves only a wallet's 10,000 most recent fills; `historyTruncated` says when that cut the window short, so very active wallets are covered only for their latest trades). Analytics only, not investment advice.

Input parameters:

- `address` (string, required): EVM 0x address.
- `curvePoints` (integer): Downsampled points in the returned PnL curve.
- `lookbackDays` (integer)

Output parameters:

- `address` (string)
- `avgHoldMinutes` (number|null)
- `avgLossUsd` (number)
- `avgR` (number|null)
- `avgWinUsd` (number)
- `closedTrades` (number)
- `disclaimer` (string)
- `fillsAnalyzed` (number)
- `historyTruncated` (boolean)
- `pnlCurve` (array)
- `pnlSharpe` (number|null)
- `style` (string)
- `topCoins` (array)
- `totalFees` (number)
- `totalRealizedPnl` (number)
- `windowDays` (number)
- `winratePct` (number)

### `hl_create_alert` (~243 tokens)

Create a standing alert

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Register a persistent alert the server evaluates on a schedule; fired events are retrieved with hl_poll_alerts. Types: funding_apr (|annualized funding| crosses a threshold), price_move (|return| over a window), whale_net_flip (cohort net position flips long/short). Turns the server into a market watcher for your agent. Analytics only, not investment advice.

Input parameters:

- `aprThreshold` (number): funding_apr: |APR| fraction (0.5 = 50%).
- `cohort` (array): whale_net_flip: up to 100 0x addresses (else HL_WHALE_ADDRESSES).
- `coin` (string, required): Perp coin symbol the alert watches.
- `cooldownMinutes` (integer): Min gap between re-fires.
- `movePct` (number): price_move: |return| fraction (0.05 = 5%, min 0.01).
- `type` (string, required)
- `windowMinutes` (integer): price_move lookback window.

Output parameters:

- `coin` (string)
- `cooldownMinutes` (number)
- `id` (string)
- `type` (string)

### `hl_list_alerts` (~60 tokens)

List your standing alerts

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] List the standing alerts registered under your API key, with their type, params, and last-fired time.

Output parameters:

- `alerts` (array)
- `count` (number)

### `hl_delete_alert` (~65 tokens)

Delete a standing alert

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Delete one of your standing alerts by id.

Input parameters:

- `id` (string, required): Alert id from hl_create_alert / hl_list_alerts.

Output parameters:

- `deleted` (boolean)

### `hl_poll_alerts` (~197 tokens)

Poll fired alerts

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Retrieve alert events that fired since your last poll. Default is simple at-most-once delivery (events are marked delivered as soon as they are read — a lost response loses them). For reliable at-least-once delivery, pass reliable=true: events are then redelivered on every poll until you acknowledge them by passing the highest event `id` you have processed as ackThroughId. Each event includes a signed signal payload verifiable with hl_signal_pubkey. Call periodically to act on the market.

Input parameters:

- `ackThroughId` (integer): Acknowledge all events with id <= this value (from a previous reliable poll).
- `limit` (integer)
- `reliable` (boolean): true = at-least-once: events redeliver until acknowledged via ackThroughId.

Output parameters:

- `acked` (number)
- `count` (number)
- `events` (array)

### `hl_signal_track_record` (~130 tokens)

Signal track record

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Transparent, verifiable performance of the signals this server has emitted: per signal type, the count, how many have been scored against forward price, hit-rate, and average/median direction-adjusted return. Every signal is cryptographically signed (verify the key via hl_signal_pubkey). Analytics only, not investment advice.

Input parameters:

- `type` (string): Filter to one signal type (e.g. 'whale_net_flip', 'funding_apr').

Output parameters:

- `note` (string)
- `records` (array)

### `hl_smart_money_score` (~258 tokens)

Smart-money screener & labels

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Rank a cohort of wallets by a composite screening score over their PAST results (risk-adjusted performance, winrate, reward/risk, capital size, activity) and tag each with behavioral labels (whale, sharp, scalper, market_maker_like, high_conviction, underwater, …). IMPORTANT: this is a descriptive screener, NOT a validated predictor — the component weights are chosen by judgement and have not been shown to predict future returns. Check `calibration` in the response for the current out-of-sample evidence, and hl_score_calibration for detail. Use it to shortlist wallets to inspect, not as a forecast. Cohort resolves from `cohort` arg, the auto-refreshed leaderboard cohort, or HL_WHALE_ADDRESSES. Analytics only, not investment advice.

Input parameters:

- `cohort` (array): 0x wallet addresses to score; else HL_WHALE_ADDRESSES.
- `limit` (integer)
- `lookbackDays` (integer)
- `offset` (integer)
- `topN` (integer)

Output parameters:

- `analyzed` (number)
- `calibration` (object)
- `cohortAgeSeconds` (number|null)
- `disclaimer` (string)
- `nextOffset` (number|null)
- `ranked` (array)
- `source` (string)

### `hl_coordination_scan` (~182 tokens)

Wallet coordination scan

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Find wallets in a cohort that hold near-identical directional exposure right now (high cosine similarity of their position vectors) — likely the same entity, a copy-bot, or a coordinated group. Returns clusters and the strongest pairs. Cohort = `cohort` arg or HL_WHALE_ADDRESSES. Analytics only, not investment advice.

Input parameters:

- `cohort` (array)
- `dex` (string): Perp dex: omit for the main dex, a HIP-3 builder dex name (e.g. 'xyz'), or 'all' for every dex (at most 20 wallets).
- `threshold` (number): Cosine similarity to link two wallets.
- `topN` (integer)

Output parameters:

- `clusters` (array)
- `dex` (string)
- `disclaimer` (string)
- `pairs` (array)
- `source` (string)
- `walletsCappedTo` (number|null)
- `walletsCompared` (number)
- `walletsQueried` (number)

### `hl_polymarket_divergence` (~269 tokens)

Polymarket ↔ Hyperliquid divergence

[PREMIUM — needs an API key: call hl_request_free_key first for a free key, or pay per call via x402] Compare Polymarket price-threshold odds for a coin (e.g. 'BTC above $100k by date') against the probability implied by Hyperliquid's price + realized volatility. Surfaces cross-market mispricings ranked by edge, checks the market's own price ladder for logically impossible orderings (P(further threshold) > P(nearer)), collapses duplicate contract listings, and reports each estimate's sensitivity to the volatility input. Rows where a diffusion model is out of its depth (extreme tails the market prices but a no-jump model cannot) are flagged, not asserted. Optionally use perp funding as drift. Estimate under lognormal dynamics. Analytics only, not investment advice.

Input parameters:

- `coin` (string, required): Perp coin symbol, e.g. 'BTC'.
- `limit` (integer)
- `minEdge` (number): Min |Polymarket − HL| probability gap to report.
- `useFundingDrift` (boolean): Use annualized funding as the drift term.
- `volInterval` (string)
- `volWindowDays` (integer): Lookback for realized volatility.

Output parameters:

- `annualizedVol` (number)
- `coin` (string)
- `disclaimer` (string)
- `divergences` (array)
- `driftUsed` (number)
- `duplicatesCollapsed` (number)
- `ladderViolations` (array)
- `marketsScanned` (number)
- `note` (string)
- `skippedImplausible` (number)
- `spot` (number)
- `volBand` (object)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersmash#diagnostics

## Score history

- 2026-11-05: 67
- 2026-11-04: 67
- 2026-11-03: 66
- 2026-11-02: 66
- 2026-11-01: 65
- 2026-10-31: 65
- 2026-10-30: 64
- 2026-10-29: 64

## Common questions

### What is the HyperSignal MCP server?

HyperSignal MCP is listed in the public MCP registry as io.github.johnsmithxy1mmm-sys/hypersignal-mcp. Hyperliquid analytics for AI agents: whale flow, liquidations, funding, risk, signed signals. This page covers its hosted endpoint (https://hypersmash.fly.dev/mcp).

### Is the HyperSignal MCP server safe to use?

HyperSignal MCP scores 67 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the HyperSignal MCP server expose?

HyperSignal MCP exposes 30 tools: hl_request_free_key, hl_buy_pro, hl_get_markets, hl_get_orderbook, hl_get_candles, and 25 more. Their descriptions and schemas cost roughly 4,277 tokens of context every time the server is loaded.

### Does the HyperSignal MCP server require authentication?

No. We connected to HyperSignal MCP without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the HyperSignal MCP server still maintained?

HyperSignal MCP is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://hypersmash.fly.dev/mcp
- Repository: https://github.com/johnsmithxy1mmm-sys/HyperliquidMCP
- Changelog RSS feed: https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersmash.xml
- Changelog JSON feed: https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersmash.json
- HTML version of this page: https://verifymcp.io/servers/johnsmithxy1mmm-sys-hypersignal-mcp/hypersmash