Skip to main content
Glama

decide

One turn of your trading agent: send a SCORE, never an order. p is your probability (0 to 1) that the market is higher at the desk's horizon, what is one line of why. Cubicle's code writes the typed decision (taifoon.trading.decision.v1) by the contract's score rule: enter long or short, stand aside, or exit (exit: true; an exit is always allowed), never stacking, inside the clock and the practice caps; every entry carries its stop and longest hold, and the code exits on them before it reads your next score. The decision's PRACTICE order goes to the devnet book (test dollars, never proof), the decision is sealed onto the coord pipe and appended to your hash-chained ledger. Returns the decision, its trace id, the order and fill, and your open position.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "market": "<market>",
      "what": "<what>"
    },
    "name": "decide"
  }
}

Input Schema

NameRequiredDescriptionDefault
pNoyour probability that the market is higher at the desk's horizon
byNoyour model or agent's name, recorded with the score (default: agent)
exitNotrue: close your open position on this market now (p is then ignored)
whatYesone line (3 to 280 characters): why you scored it so
marketYesthe market: BTC, ETH, SOL (crypto clock) or NQ, ES (CME sessions)
stop_bpsNoa tighter stop than the desk's, in basis points (the desk's is the ceiling)
max_hold_sNoa shorter longest hold than the desk's, in seconds (the desk's is the ceiling)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.2/5.0
Behavior5/5

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

Despite annotations declaring a non-read-only, open-world, non-idempotent write, the description adds substantial behavior: the code applies contract score rules, enforces practice caps and the clock, attaches stops and max holds and exits on them before the next score, routes only a PRACTICE order to the devnet book ("test dollars, never proof"), and seals the decision onto the coord pipe and hash-chained ledger. That is exactly the side-effect and safety context an agent needs before a mutation.

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

Conciseness3/5

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

The key instruction is front-loaded well, but the body is a dense semicolon-chained run-on mixing parameter semantics, engine rules, settlement, and return values. Several clauses ("what is one line of why", "p is your probability (0 to 1)") duplicate the schema verbatim and could be dropped.

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?

For a write tool with no output schema, the description covers the return payload (decision, trace id, order and fill, open position) and the full side-effect chain. It falls short only on failure behavior: what happens when a score arrives outside the clock or breaching practice caps is alluded to but never stated.

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 all seven parameters including p, what, exit, stop_bps and max_hold_s are already documented. The description restates p's meaning and the exit semantics, and notes entries carry a stop and longest hold, but adds almost nothing the schema doesn't already say about parameter format or bounds.

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?

"One turn of your trading agent: send a SCORE, never an order" states the verb, the resource (a score), and the critical distinction from order-placement. It then enumerates the exact decision outcomes the tool produces (long, short, stand aside, exit), so an agent knows precisely what calling this accomplishes.

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?

"One turn of your trading agent" establishes the cadence and context for calling, and the exit clause ("exit: true; an exit is always allowed") plus "never stacking" give clear conditional guidance. What's missing is an explicit pointer to alternatives such as the sibling `decisions` or `desk_read` for inspection rather than submission.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources