Hyperliquid Agent Gateway
PYPI · HYPERLIQUID-AGENT-GATEWAY · SCANNED OCT 1
Read-only MCP gateway to Hyperliquid: liquidation distance, funding carry, HyperEVM flows. No keys.
Available components
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security50
- Malware scan not yet available for this package.Unverified
- No known CVEs affecting this package version or its production dependencies.Pass
- Runs hatchling.build at install time, a recognised build step with no custom scripting around it. View diagnostics → Pass
- 1 of 15 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency97
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Cryptographically verified build provenance (signed, bound to alekskram/hyperliquid-agent-gateway). View diagnostics → Pass
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 8 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability64
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 1624 tokens (~135/item across 12 items; 12 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management30
- Stability observed for 9 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage71
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 0% of tool parameters carry a description.Fail
- Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 12 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 13 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a current MCP spec version (2026-07-28).Pass
How do I install the Hyperliquid Agent Gateway MCP server?
Hyperliquid Agent Gateway runs locally as a PyPI package, launched with uvx hyperliquid-agent-gateway. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
pypi · hyperliquid-agent-gateway
claude mcp add alekskram-hyperliquid-agent-gateway -- uvx hyperliquid-agent-gateway
{
"mcpServers": {
"alekskram-hyperliquid-agent-gateway": {
"command": "uvx",
"args": [
"hyperliquid-agent-gateway"
]
}
}
} {
"servers": {
"alekskram-hyperliquid-agent-gateway": {
"command": "uvx",
"args": [
"hyperliquid-agent-gateway"
]
}
}
} codex mcp add alekskram-hyperliquid-agent-gateway -- uvx hyperliquid-agent-gateway
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"alekskram-hyperliquid-agent-gateway": {
"type": "local",
"command": [
"uvx",
"hyperliquid-agent-gateway"
],
"enabled": true
}
}
} openclaw mcp add alekskram-hyperliquid-agent-gateway --command uvx --arg hyperliquid-agent-gateway
mcp_servers:
alekskram-hyperliquid-agent-gateway:
command: "uvx"
args: ["hyperliquid-agent-gateway"] {
"McpServers": {
"alekskram-hyperliquid-agent-gateway": {
"Transport": "stdio",
"Command": "uvx",
"Arguments": [
"hyperliquid-agent-gateway"
]
}
}
} assistant mcp add alekskram-hyperliquid-agent-gateway -t stdio -c uvx -a hyperliquid-agent-gateway
{
"mcpServers": {
"alekskram-hyperliquid-agent-gateway": {
"command": "uvx",
"args": [
"hyperliquid-agent-gateway"
]
}
}
} Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 1 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.
- 30 Sept 26 −15
- Malware scan: pass → unverified ▼ security
- 29 Sept 26 +16
- Malware scan: unverified → pass ▲ security
- 28 Sept 26 −5
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
- 27 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.
- 25 Sept 26 +1
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
- 23 Sept 26 +1
- Tool safety: pass → unverified ▼ security
- Stability: Stability not yet verified: we do not have a sandbox capture of the MCP schema this version of the package serves yet. security
- Capabilities: pass → unverified ▼ functional
- Tool coverage: 100 → unverified ▼ functional
- Stability: unverified → 0.03 ▲ functional
- First check of Schema quality: unverified functional
- Package version: 0.1.2 → 0.1.3 functional
- 22 Sept 26 66
First indexed and scored.
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 1 Oct 2026 · Analysed pypi/hyperliquid-agent-gateway@0.1.3
Provenance Verified
A signed build attestation was found and verified, binding this exact artifact to the source repository it claims to come from.
| Result | Verified |
|---|---|
| Ecosystem | pypi |
| Reason | Verified |
| Discovered via | Registry attestation endpoint |
| Source repo | alekskram/hyperliquid-agent-gateway |
| Certificate issuer | https://token.actions.githubusercontent.com |
| Certificate SAN | https://github.com/alekskram/hyperliquid-agent-gateway/.github/workflows/release.yml@refs/tags/v0.1.3 |
| Rekor log index | 2910453744 |
| Predicate type | PyPI publish attestation https://docs.pypi.org/attestations/publish/v1 |
| Subject digest | sha256:6291b34d155017c202094b4d183b1654f83955629d264af1d61265d1af7ec231 |
Background: How many MCP packages publish verified provenance →
Install scripts 1 script
| Hook | Tier | Command |
|---|---|---|
| build_backend | allowlisted | hatchling.build |
Background: Why install scripts are a supply-chain risk →
Dependencies 15 packages
| Packages resolved | 15 |
|---|---|
| Stale | 1 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
candles Candles ~119
OHLCV candles from candleSnapshot for `coin`: rows newest-first, capped at `limit` (max 500). startTime is computed so roughly `limit` candles of `interval` are requested. Valid intervals: 1m, 15m, 1h, 4h, 1d, 1w, 1M. Example: candles(coin="BTC", interval="4h", limit=50)
| Name | Type | Req | Description |
|---|---|---|---|
| coin | string | yes | – |
| interval | string | – | – |
| limit | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
funding_carry_screener Funding Carry Screener ~119
Rank ALL perps from ONE metaAndAssetCtxs call by `metric` (premium default; oi / volume options) and return the topN rows with premium, oi, dayVolume, markPx. fundingHistory rows are fetched ONLY for the topN coins (weight economy) and appear as 'funding_history' on rows where that call succeeded. Example: funding_carry_screener(topN=5, metric="premium")
| Name | Type | Req | Description |
|---|---|---|---|
| metric | string | – | – |
| topN | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
funding_history Funding History ~94
Hourly funding history for `coin` (fundingHistory) plus the CURRENT premium from metaAndAssetCtxs (premium_now): rows carry fundingRate (hourly), premium, time (ms). Newest-first, capped at `limit` (max 500). Example: funding_history(coin="BTC", limit=48)
| Name | Type | Req | Description |
|---|---|---|---|
| coin | string | yes | – |
| limit | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
liquidation_risk Liquidation Risk ~239
KEY TOOL. Liquidation risk for one perp account (clearinghouseState + userFunding): margin summary (accountValue, totalRawUsd, totalMarginUsed, withdrawable), cross maintenance margin, per-position leverage/entry/mark/unrealizedPnl and liquidationPx when the venue provides it. Positions in the LIVE clearinghouseState response carry NO markPx: mark is resolved per coin from metaAndAssetCtxs (fallback allMids) and each row says where it came from in mark_px_source ('position' | 'metaAndAssetCtxs' | 'allMids' | null). When liquidationPx is null, liq_distance_pct is an ESTIMATE from the maintenance-margin ratio (crossMaintenanceMarginUsed / marginSummary.accountValue; isolated positions use marginUsed / positionValue) - flagged 'estimated': true, liqPx stays null (honest). funding_drag totals negative funding payments from userFunding. Address must be 0x + 40 hex. Example: liquidation_risk(address="0x...")
| Name | Type | Req | Description |
|---|---|---|---|
| address | string | yes | – |
Structured output declared, but exposes no named fields.
No examples provided.
market_overview Market Overview ~112
Perp market overview from ONE metaAndAssetCtxs call (~233 perps): per-coin mark, open interest, day volume, premium (funding basis), max leverage, plus totals (sum OI, sum day volume). Rows are sorted by `sort` in {open_interest (default), volume, premium} descending and capped at `limit` (max 100). Example: market_overview(limit=10)
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | – |
| sort | string | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
order_book Order Book ~97
Aggregated order book for one coin from l2Book: `depth` levels per side (default 10, max 100), aggregated by the book's nSigFigs precision (default taken from the response levels), with total liquidity (sum px*sz) per side. Example: order_book(coin="ETH", depth=20)
| Name | Type | Req | Description |
|---|---|---|---|
| coin | string | yes | – |
| depth | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
quote Quote ~125
One coin's live quote from l2Book: bid/ask/mid/spread and top-of-book sizes. mid is the book midpoint ((bid+ask)/2) whenever both sides are present ('book'); allMids is only a labeled fallback when the book lacks a side ('allMids'), so mid never escapes the [bid, ask] range. The coin is validated against the perp universe (spot '@{index}/NAME' pairs pass through to the book). Example: quote(coin="BTC")
| Name | Type | Req | Description |
|---|---|---|---|
| coin | string | yes | – |
Structured output declared, but exposes no named fields.
No examples provided.
spot_overview Spot Overview ~106
Spot market overview (~326 pairs) from spotMeta + spotMetaAndAssetCtxs: pair rows (name, mark, day volume, OI where published) with HIP-1 <-> ERC-20 links resolved - '@{index}' spot coin names are mapped to token names via spotMeta.tokens. Sorted by day volume descending, capped at `limit` (max 100). Example: spot_overview(limit=10)
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
token_transfers Token Transfers ~181
Recent ERC-20 Transfer events for `contract` on HyperEVM via eth_getLogs with an ADAPTIVE window (starts 30 blocks, halves on too-big/failing responses, doubles on quiet ones up to 60; active tokens need ~30-60 block windows). Rows (newest-first, max 500): from, to, value (raw/1e18, 6dp), txHash, blockNumber, ts when the log carries a timestamp. `from_block` anchors the walk (default: latest - 300). Errors degrade to an honest error dict naming the RpcError kind (rate-limit vs window-too-wide narrative). Example: token_transfers(contract="0x...", limit=50)
| Name | Type | Req | Description |
|---|---|---|---|
| contract | string | yes | – |
| from_block | – | – | – |
| limit | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
trader_activity Trader Activity ~111
Trading profile for one address (userFills + userFunding + clearinghouseState): total closed PnL, fees (fee + builderFee), volume, fill count, win-rate-ish share of fills with closedPnl > 0, per-coin breakdown, net funding and open positions. `limit` caps fills analyzed (max 500, newest-first). Example: trader_activity(address="0x...", limit=100)
| Name | Type | Req | Description |
|---|---|---|---|
| address | string | yes | – |
| limit | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
trades Trades ~78
Recent public fills from recentTrades(coin) WITH both sides' addresses: rows (newest-first, max 500) carry side, px, sz, time, coin, hash and users [maker, taker]. Example: trades(coin="BTC", limit=10)
| Name | Type | Req | Description |
|---|---|---|---|
| coin | string | yes | – |
| limit | integer | – | – |
Structured output declared, but exposes no named fields.
No examples provided.
wallet_balance Wallet Balance ~116
Wallet balances on HyperEVM + Hyperliquid spot: native balance (eth_getBalance), ERC-20 balances via eth_call balanceOf for up to 20 tokens resolved from spotMeta tokens with 0x contract addresses (cap respects the 100 req/min rpc budget), and Hyperliquid spot balances via spotClearinghouseState. Per-token rows: token, balance_raw (wei), balance (units, 6dp), source. Example: wallet_balance(address="0x...")
| Name | Type | Req | Description |
|---|---|---|---|
| address | string | yes | – |
Structured output declared, but exposes no named fields.
No examples provided.
What is the Hyperliquid Agent Gateway MCP server?
Hyperliquid Agent Gateway is an MCP server listed in the public MCP registry as io.github.alekskram/hyperliquid-agent-gateway. Read-only MCP gateway to Hyperliquid: liquidation distance, funding carry, HyperEVM flows. No keys. This page covers its PyPI package (hyperliquid-agent-gateway).
Is the Hyperliquid Agent Gateway MCP server safe to use?
Hyperliquid Agent Gateway scores 66 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 1 October 2026. Its build provenance is signed and verified. 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 Hyperliquid Agent Gateway MCP server expose?
Hyperliquid Agent Gateway exposes 12 tools: market_overview, spot_overview, quote, order_book, candles, and 7 more. Their descriptions and schemas cost roughly 1,497 tokens of context every time the server is loaded.
Is the Hyperliquid Agent Gateway MCP server still maintained?
Hyperliquid Agent Gateway is still listed as active in the MCP registry. We last reached this channel on 1 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the Hyperliquid Agent Gateway MCP server under?
Hyperliquid Agent Gateway declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.