Skip to main content
Glama

aeX402 — Cross-Chain DeFi MCP: LINQ, AMM, Bridge, AI

by com.aex402
OverviewDiscussionsAdmin

hl_info

Query the Hyperliquid Info API — the perps/spot order-book data layer (read-only, no auth). The 'type' field selects the query: meta (perps universe + leverage), l2Book (order book for a coin), allMids (mid prices), clearinghouseState (a user's positions/margin), spotMeta, meta+assetCtxs (mark/funding/OI), candleSnapshot, and more. For type:metaAndAssetCtxs specifically: the raw response is TWO PARALLEL top-level arrays (universe names, asset contexts) correlated only by array position — do not try to read/scan it directly for '200+ coins' style questions (funding-rate outliers, biggest open interest, highest volume). Pass rankBy instead: it joins the two arrays server-side, sorts, and returns only the top N — the correct way to answer 'what has the highest/most extreme funding/OI/volume right now'.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
dirNorankBy only: asc | desc (default desc — highest rankBy value first)
coinNoCoin symbol for l2Book/candleSnapshot (e.g. BTC, ETH)
typeYesInfo request type, e.g. meta, l2Book, allMids, clearinghouseState, metaAndAssetCtxs
userNo0x address for user-scoped queries (clearinghouseState, openOrders, userFills)
limitNorankBy only: how many ranked rows to return (default 20, max 100)
rankByNotype:metaAndAssetCtxs ONLY. One of funding | fundingAbs | openInterest | volume. Joins each coin's name to its funding/OI/volume, drops delisted coins, sorts, and returns {ranked, rankBy, dir, totalCoins} instead of the raw two-array response. Use fundingAbs for "biggest funding outlier either direction" (ranks by |funding|); use funding for "most positive/most negative" specifically (with dir).
endTimeNocandleSnapshot only: end time epoch ms (default: now)
intervalNocandleSnapshot only: candle interval, e.g. 1m,5m,15m,1h,4h,1d
startTimeNocandleSnapshot only: start time epoch ms (default: endTime minus 500 intervals)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changed
    • addedInput schema / properties / endTime
      Added value: +{
+  "description": "candleSnapshot only: end time epoch ms (default: now)",
+  "type": "number"
+}
    • addedInput schema / properties / interval
      Added value: +{
+  "description": "candleSnapshot only: candle interval, e.g. 1m,5m,15m,1h,4h,1d",
+  "type": "string"
+}
    • addedInput schema / properties / startTime
      Added value: +{
+  "description": "candleSnapshot only: start time epoch ms (default: endTime minus 500 intervals)",
+  "type": "number"
+}
  2. Changed3 schema fields changed
    • addedInput schema / properties / dir
      Added value: +{
+  "description": "rankBy only: asc | desc (default desc — highest rankBy value first)",
+  "type": "string"
+}
    • addedInput schema / properties / limit
      Added value: +{
+  "description": "rankBy only: how many ranked rows to return (default 20, max 100)",
+  "type": "number"
+}
    • addedInput schema / properties / rankBy
      Added value: +{
+  "description": "type:metaAndAssetCtxs ONLY. One of funding | fundingAbs | openInterest | volume. Joins each coin's name to its funding/OI/volume, drops delisted coins, sorts, and returns {ranked, rankBy, dir, totalCoins} instead of the raw two-array response. Use fundingAbs for \"biggest funding outlier either direction\" (ranks by |funding|); use funding for \"most positive/most negative\" specifically (with dir).",
+  "type": "string"
+}
  3. First observed

TDQS

A4.6/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden of behavioral disclosure. It states read-only and no auth, and explicitly warns about the two-parallel-array structure and its pitfalls. It also describes the rankBy response format. It does not mention rate limits or error behavior, but for a query tool with a well-defined API, this is a minor gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is fairly long but every sentence contributes. The front-loaded purpose and the prominent warning about the metaAndAssetCtxs pitfall make it structured and effective. It could be tightened slightly, but the density of information justifies the length.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity (9 parameters, no output schema), the description covers the tricky aspects: the rankBy alternative, parameter scoping per type, and the specific use cases. It does not detail error handling or rate limits, but for a query tool, the description is sufficient to invoke correctly. Minor omissions prevent a 5.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Although the schema already documents all parameters (100% coverage), the description adds substantial value by explaining the semantics of rankBy in detail (funding vs fundingAbs, dir, and the server-side join behavior). It also clarifies the relationship between startTime and endTime for candleSnapshot and the default for limit. This goes well beyond the schema descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool queries the Hyperliquid Info API, identifies it as the perps/spot order-book data layer, and lists the query types it supports. It distinguishes itself from the sibling tools by naming the specific API and its purpose, which is unambiguous even without comparing to siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly explains when to use rankBy instead of the raw two-array response, warns against scanning the raw response for ranking questions, and specifies the conditions for using funding vs fundingAbs. It also clarifies which parameters apply to which query types (e.g., coin for l2Book/candleSnapshot, user for clearinghouseState). This gives clear context and exclusions, going beyond a generic usage note.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All aeX402 — Cross-Chain DeFi MCP: LINQ, AMM, Bridge, AI toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources