TradeHub

Close a position (reduce-only)

close_perp_position_reduce_only

Destructive

Close all or part of a perpetual futures position with a reduce-only market order, so it can never open or flip a position. Use to take profit, cut a loss or reduce exposure; it is also the only way to undo a filled order. Third-party apps can close only positions they opened. Not for cancelling resting orders (use cancel_open_order). Inputs: symbol, venue, close_pct (1-100, default 100). Works while trading is halted. Next: get_my_open_positions.

Input Schema

Name	Required	Description	Default
`venue`	No	Venue: sandbox, hyperliquid, okx, bybit or binance. Omitted: the agent's only venue, else hyperliquid, else your only connected venue, else sandbox	hyperliquid
`symbol`	Yes	Base coin, e.g. BTC
`close_pct`	No	Percent of the position to close, 1-100

Output Schema

Name	Required	Description
`ok`	Yes	False when rejected or failed
`fee`	No	Once executed: the trading fee, estimated at once and replaced by the amount the exchange charged once its fill records show it
`legs`	No	A ladder entry: per leg {price, size_base, order_id, sl_oid, status}; status resting, or after a refused leg withdrawn, filled or partly_filled
`error`	No	Why it was rejected or failed
`venue`	No	Venue name
`checks`	No	Checks run at execution
`sl_oid`	No	Venue order id of the protective stop
`status`	Yes	pending_approval, armed (a trigger entry waiting for its price), executed, rejected, failed, expired, declined, cancelled, or executing (TradeHub has not confirmed the outcome yet: check get_order_request_status)
`receipt`	No	Signed receipt (compact JWS, EdDSA) once executed: request, checks, preview price, fill, slippage, order ids, delegation. Verify with /.well-known/tradehub-receipts.jwks
`order_id`	No	Venue order id of the fill
`fill_price`	No	Average fill, USD per coin
`request_id`	Yes	Order request id; get_order_request_status(request_id) follows it
`deprecation`	No	Present only while this tool is deprecated: replacement, removal_date, changelog
`filled_size`	No	Filled size, in the base coin (e.g. BTC), not USD
`take_profits`	No	set_take_profit_levels: the take-profits now on the exchange
`trigger_price`	No	status armed: opens at market when the mark reaches this, USD per coin
`trigger_expires_at`	No	status armed: when it lapses unless reached, ISO 8601 UTC
`approval_expires_at`	No	While pending: when the approval request lapses, ISO 8601 UTC

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 29, 2026

addedOutput schema / properties / legs

Added value: +{
+  "anyOf": [
+    {
+      "items": {
+        "additionalProperties": true,
+        "type": "object"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "A ladder entry: per leg {price, size_base, order_id, sl_oid, status}; status resting, or after a refused leg withdrawn, filled or partly_filled",
+  "title": "Legs"
+}

Changed2 schema fields changedOct 27, 2026

addedOutput schema / $defs / FeeOut

Added value: +{
+  "additionalProperties": true,
+  "properties": {
+    "actual": {
+      "anyOf": [
+        {
+          "type": "number"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "The charged amount in asset; negative = a rebate",
+      "title": "Actual"
+    },
+    "actual_usd": {
+      "anyOf": [
+        {
+          "type": "number"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "What the exchange charged, USD; null until its fill records show it (or when charged in another asset, see actual and asset)",
+      "title": "Actual Usd"
+    },
+    "asset": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "The asset it was charged in (USDT, USDC, BNB …)",
+      "title": "Asset"
+    },
+    "deprecation": {
+      "anyOf": [
+        {
+          "additionalProperties": true,
+          "type": "object"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "description": "Present only while this tool is deprecated: replacement, removal_date, changelog",
+      "title": "Deprecation"
+    },
+    "estimated_usd": {
+      "anyOf": [
+        {
+          "type": "number"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "The account's rate x the filled notional, USD",
+      "title": "Estimated Usd"
+    },
+    "kind": {
+      "description": "open or close",
+      "title": "Kind",
+      "type": "string"
+    },
+    "status": {
+      "description": "pending (not read yet), done, or unknown (the exchange never showed it)",
+      "title": "Status",
+      "type": "string"
+    }
+  },
+  "required": [
+    "kind",
+    "estimated_usd",
+    "actual_usd",
+    "actual",
+    "asset",
+    "status"
+  ],
+  "title": "FeeOut",
+  "type": "object"
+}

addedOutput schema / properties / fee

Added value: +{
+  "anyOf": [
+    {
+      "$ref": "#/$defs/FeeOut"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Once executed: the trading fee, estimated at once and replaced by the amount the exchange charged once its fill records show it"
+}

First observedOct 27, 2026

TDQS

A4.7/5.0

Behavior5/5

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

Annotations already declare destructive/non-idempotent/openWorld, but the description adds context they do not carry: the reduce-only guarantee against flipping, the third-party scoping restriction, and that the tool still functions while trading is halted. It also states the follow-up ('Next: get_my_open_positions'), which is useful operational sequencing for a destructive tool.

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?

Dense but well ordered: what it does, why use it, what it is not, constraints, inputs, next action. Every sentence carries information an agent would otherwise have to guess.

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?

An output schema exists so return values need not be explained; the description instead covers the gaps that matter for a destructive market-order tool: scope of effect, restrictions, availability during halts, and the preferred follow-up call.

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 coverage is 100%, so venue, symbol and close_pct are already fully documented. The description's 'Inputs: symbol, venue, close_pct (1-100, default 100)' largely restates the schema rather than adding new meaning, which is the expected baseline-3 outcome when the schema does the heavy lifting.

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 + mechanism: 'close all or part of a perpetual futures position with a reduce-only market order'. The scope guarantees ('can never open or flip a position') and the explicit sibling exclusion ('Not for cancelling resting orders (use cancel_open_order)') let an agent separate it from cancel_open_order and amend_open_entry_order immediately.

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?

Names concrete use cases (take profit, cut a loss, reduce exposure, undo a filled order), an explicit when-not case routed to a named sibling, and a permissions constraint ('third-party apps can close only positions they opened'). Preconditions and availability are covered ('works while trading is halted'), leaving little to inference.

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

All TradeHub 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?