AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

Desk Operating Model

A constitution for running multiple AI bots as a single trading desk, with defined seats, an evidence standard, and a hard approval gate.

AutomationAdvanced★ 372⑂ 85AI score 8/10Last updated: Nov 4, 2026

What it does

  • Seats and roles: Defines seven bots — Desk Lead, Market Analyst, Research Analyst, Strategist, Risk Manager, Execution Trader, Trade Reviewer — and exactly who may write to the exchange (only the Execution Trader).
  • Shared workspace layout: Specifies the /workspace/trading-desk/ file conventions for proposals, briefs, research dossiers, strategy labs, and the journal, and keeps secrets out of the workspace entirely.
  • Engagement levels: Research desk → testnet → mainnet, with level changes reserved for the user and recorded in desk.md.
  • Evidence standard: Every number carries a source, network, and UTC timestamp; unverifiable data is "unavailable", not a quiet negative; bot agreement never counts as evidence.
  • Approval model: Orders ship only after the user types the ticket id; bots may never type, paste, relay, or simulate an approval, and the transcript is not proof.
  • Handoff format: A fixed short block carrying proposal id, ask, evidence, constraints, deliverable, and next owner.

Who it's for

  • Builders standing up an agent-based trading desk on Hyperliquid.
  • Anyone splitting work across many agents while tightly fencing order-execution authority.
  • Teams looking for a reusable pattern for approval gates and responsibility separation in multi-agent systems.

Example uses

  1. Bootstrapping a desk and recording network, account, engagement level, and standing instructions in desk.md.
  2. Passing a signal from the Market Analyst to the Risk Manager with a HG-20260816-01 | to: @Risk Manager handoff block.
  3. Triaging an ambiguous request via the three questions: who owns the outcome, what evidence settles it, and does it touch the exchange write path.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/galleonlabs/hypergrok-trading-desk/HEAD/skills/desk-operating-model/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the skills/desk-operating-model folder from the GitHub repo galleonlabs/hypergrok-trading-desk into my ~/.claude/skills/desk-operating-model/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/galleonlabs/hypergrok-trading-desk.git && mkdir -p ~/.claude/skills && cp -r hypergrok-trading-desk/skills/* ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repository: git clone https://github.com/galleonlabs/hypergrok-trading-desk.git
  3. Create the skills directory: mkdir -p ~/.claude/skills
  4. Copy the skills in: cp -r hypergrok-trading-desk/skills/* ~/.claude/skills/
  5. Confirm ~/.claude/skills/desk-operating-model/SKILL.md exists.
  6. Restart Claude Code and ask something like "explain the desk operating model" to verify the skill loads.
  7. Before touching real money, start at level 1 (research desk) and only move to testnet/mainnet after verifying that a native approval gate covers every exchange write route (SDK, scripts, direct HTTP).
View source on GitHub ↗License: MIT