whale-data

Recorded liquidations for one coin

coin_liquidations

Read-onlyIdempotent

Forced closes ALREADY recorded for one pair, split long versus short. These are executed liquidations, not projected levels: for where liquidations could happen next use liq_zones, and for the whole market at once use market_liquidations. The plain-English name for liquidations (identical data and credit cost). Descriptive market data only.

Input Schema

TableJSON Schema

Name	Required	Description
`coin`	No	Alias for pair.
`pair`	No	e.g. ETH/USD. A bare symbol like ETH also works.
`hours`	No	Look back this many hours, up to 720. Omit for the most recent events.
`limit`	No	Max rows returned. Capped at 25 on the free plans and 1000 on a paid plan; asking for more returns the cap, and the payload says so in `capped`. count and the *_usd totals always cover the whole window regardless.
`since_ms`	No	Polling cursor: only rows newer than this Unix-ms timestamp, applied before the row cap. Pass back the next_since_ms from the previous answer so a timer never re-reads the same rows.

Output Schema

TableJSON Schema

Name	Required	Description
`note`	No	One line saying what the numbers cover, or why there are none. Read it before reporting any zero as a market reading.
`pair`	No	The pair these events belong to.
`count`	No	How many events fired in the whole window, not how many rows came back.
`capped`	No	Present only when an argument was cut down to a ceiling. Absent means nothing was clamped, so a smaller number than you asked for is the data and not a limit.
`sources`	No	How the events split between the streamed exchanges and Hyperliquid, and whether the Hyperliquid part is a sample.
`long_usd`	No	Dollars of longs force-closed across the window.
`available`	No	False when this lane could not answer: a producer is dark, the pair sits outside this plan, or the argument named something we do not carry. Absent or true means the numbers are a real reading, not a guess.
`short_usd`	No	Dollars of shorts force-closed across the window.
`truncated`	No	True when more rows matched than were returned, so what you have is a page and not the whole set.
`access_note`	No	What this connection could not return and what lifts it. Always pass this on to the user: it is the only place that information appears.
`liquidations`	No	The events themselves, newest first: price, size in base and quote, side (long_liquidated or short_liquidated), exchange and time. A sample of the window when truncated is true.
`window_hours`	No	The period the totals cover, in hours.
`rows_returned`	No	How many event rows are actually in this reply.
`window_minutes`	No	The same period in minutes.
`requested_hours`	No	The period asked for, so a page can be told apart from a period.
`total_in_window`	No	How many events the window holds in total.
`history_included`	No	True when the recorded series was merged in, so a restart has not erased earlier events.
`total_volume_usd`	No	Dollars forced out across the whole window.
`long_liquidations`	No	COUNT of longs force-closed. For dollars read long_usd.
`rows_span_minutes`	No	How far back the returned rows themselves reach.
`short_liquidations`	No	COUNT of shorts force-closed. For dollars read short_usd.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedSep 15, 2026

changedInput schema / properties / limit / description

Previous value: -"Max rows returned. Capped at 25 keyless and 1000 on a paid plan; asking for more returns the cap, and the payload says so in `capped`. count and the *_usd totals always cover the whole window regardless."New value: +"Max rows returned. Capped at 25 on the free plans and 1000 on a paid plan; asking for more returns the cap, and the payload says so in `capped`. count and the *_usd totals always cover the whole window regardless."

Changed1 schema field changedSep 5, 2026

addedInput schema / properties / since_ms

Added value: +{
+  "description": "Polling cursor: only rows newer than this Unix-ms timestamp, applied before the row cap. Pass back the next_since_ms from the previous answer so a timer never re-reads the same rows.",
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}

AddedSep 2, 2026

TDQS

A4.5/5.0

Behavior4/5

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

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, establishing a safe read-only operation. The description adds valuable context beyond these annotations: it clarifies these are executed liquidations (not projected levels), split by long/short, and characterizes the data as descriptive market data only. This enriches the agent's understanding of the tool's behavior without contradicting the annotations.

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

Conciseness5/5

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

The description is compact and front-loaded with the core purpose, then supplies crucial sibling differentiation and a note about the tool being the plain-English name for liquidations. Each sentence adds distinct value—purpose, behavioral distinction, and routing—with no filler or redundant repetition of schema details.

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

Completeness5/5

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

Given that an output schema exists and annotations already cover safety and idempotency, the description addresses everything an agent needs to correctly select and invoke the tool: what it returns (recorded, split long/short), when to use it versus alternatives, and the nature of the data (descriptive, not projected). No critical context appears missing.

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

Parameters3/5

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

Schema description coverage is 100%, so the input schema fully documents all five parameters (coin, pair, hours, limit, since_ms) including their semantics and constraints. The description does not add any parameter-specific meaning beyond what the schema already provides, so it meets the baseline of 3 without exceeding it.

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 states exactly what the tool does: returns already-recorded forced closes (liquidations) for one pair, split into long versus short. It distinguishes itself from sibling tools liq_zones (projected levels) and market_liquidations (whole market) explicitly, so an agent can tell them apart without opening schemas.

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?

It names specific alternatives and the conditions for selecting them: use liq_zones for where liquidations could happen next, and market_liquidations for the whole market at once. It also clarifies it is for recorded data on a single pair, leaving no ambiguity about when to invoke this tool.

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

All whale-data tools Inspect

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

Need Help?