AgentFeed

Liquidation cascade forecast

get_cascade_forecast

Read-onlyIdempotent

Liquidation forecast, FORWARD-LOOKING and not a description of what already happened. Returns the probability that a symbol will liquidate more in the NEXT 15 minutes than its own 90th-percentile 15-minute window. Calibrated on our whole multi-venue liquidation tape, over 6 million liquidations, refit every 6 hours; no exchange publishes that history, so it cannot be rebuilt. Every answer carries the exact question, the threshold in USD, the window it read, the number of historical occurrences behind the number, and instructions for settling it yourself from the public feed. When a state has too little history the tool DECLINES rather than guessing, and says why. The settled record is free at /api/forecast-record. Costs 0.02 USDC per call (x402, USDC on Solana, Base, Polygon, Arbitrum or Arc).

Input Schema

Name	Required	Description	Default
`symbol`	No	SOL, BTC, ETH or any USDT perp e.g. SXTUSDT (default SOL)
`symbols`	No	comma separated for a batch, max 20, e.g. SOL,BTC,ETH

Output Schema

Name	Required	Description	Default
`data`	Yes	the get_cascade_forecast payload; a real captured example is free at https://x402.ochinimus.app/api/sample/get_cascade_forecast
`tool`	Yes	the tool that produced this payload

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedSep 25, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": false,
+  "properties": {
+    "data": {
+      "additionalProperties": {},
+      "description": "the get_cascade_forecast payload; a real captured example is free at https://x402.ochinimus.app/api/sample/get_cascade_forecast",
+      "propertyNames": {
+        "type": "string"
+      },
+      "type": "object"
+    },
+    "tool": {
+      "const": "get_cascade_forecast",
+      "description": "the tool that produced this payload",
+      "type": "string"
+    }
+  },
+  "required": [
+    "tool",
+    "data"
+  ],
+  "type": "object"
+}

First observedSep 6, 2026

TDQS

A4.4/5.0

Behavior5/5

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

Annotations only cover the safety profile (readOnly, idempotent, non-destructive, open-world). The description adds substantial non-structured context: 0.02 USDC per call via x402 on named chains, decline-rather-than-guess behavior with a stated reason, refit cadence, calibration provenance, and self-settlement instructions. This is well beyond what annotations convey.

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 single differentiating claim (forward-looking, next-15-min probability) is front-loaded, and each subsequent sentence carries useful content (return contents, decline behavior, cost). It is dense and longer than ideal, but little is filler.

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?

With an output schema present, the description need not explain return values, and it still covers the decision-critical extras: cost, decline semantics, calibration, and settlement. An agent has everything needed to decide and invoke correctly.

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% and both parameters (symbol, symbols) are documented there, so the baseline is 3. The description references 'the window it read' and 'the threshold in USD' but adds no syntax, batch limits, or default detail beyond the schema.

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?

States a specific verb+resource (forward-looking liquidation forecast) and pins down the exact output semantics: probability a symbol liquidates more in the next 15 minutes than its own 90th-percentile window. The explicit 'FORWARD-LOOKING and not a description of what already happened' cleanly separates it from sibling history tools like get_cascade_history.

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

Usage Guidelines4/5

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

Contrasts the tool against retrospective alternatives ('not a description of what already happened') and tells the agent about the free settled record at /api/forecast-record, which implies the paid-vs-free choice. However it never names the sibling get_cascade_forecast_free or get_cascade_alert directly, so routing is inferred rather than explicit.

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

All AgentFeed 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.