Run crypto futures trades with an agent under trader-set rules
A trader connected OpenClaw to OpenAlgo and a crypto futures exchange; it made 47 trades, 4 manually overridden including a rogue ETH long, then execution permission was removed. ·
Done withOpenClaw
- What they did
- The trader ran OpenClaw on a Mac Mini M4, with OpenAlgo handling order placement and position management on live exchange feeds. They wrote their momentum strategy as plain-language instructions. Stop losses and a daily loss kill switch sat outside the agent. Exchange keys went through the broker's official system. Setup took about an hour. The trader monitored throughout and overrode some trades.
- How it went
- Over 14 days: +4.8% net, 47 trades, 57% win rate, -6.2% max drawdown, 4 manual overrides. A skill update changed how a condition was read, producing the rogue ETH long. The trader then limited the agent to research and signals.
- Worth knowing
- Keep stop losses and kill switches outside the agent, and audit third-party skills. One ClawHub skill was sending unauthorised outbound calls, so the trader revoked and regenerated the API keys.
Try it yourself with OpenClaw
Connect to my [broker or exchange] account in paper trading mode and trade [instrument] only under these rules: max position size [amount], max loss per trade [percent], allowed hours [hours], no new positions after [daily loss limit]. Before placing any order, show me the order details and wait for my yes. At the end of each day, send me a log of every trade and any rule you came close to breaking.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: mexc.com · Nov 10, 2026 · Submitted by from search gap: ｔｒａｄｅ
Related use cases
Manage 18 agents from a Notion workspace
Notion used as the control plane for a fleet of 18 OpenClaw agents.
Triage failed GitHub Actions runs and self-heal flaky tests
A QA engineer's OpenClaw agent reads failed CI logs, classifies failures, posts analysis to GitHub and Slack and opens issues; flakiness fell from 8% to 0.8%, though it once wrongly altered an assertion.
Build and maintain a local ops dashboard with agents
A solo business owner had OpenClaw agents build and iterate on a local mission-control dashboard with Kanban, approvals, cron and social scheduling, replacing Asana; open-ended overnight runs burned tokens.
Submitted by @jennabrinning
Shop for umbrella insurance through your personal agent
Personal AI agents use Coverage Cat's Agent API/MCP to size umbrella coverage and compare carrier options with price ranges, backed by a licensed broker.
Custom agent
Generate investment analysis from live market data
The poster built a personal wealth advisor in Claude Code that feeds live market data, macro indicators and news into investment analysis.
Find unclaimed money across old addresses
A user's AI agent searched prior addresses and found a further $500 in unclaimed money, then filled out forms and prepped ID checks.
Custom agent
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.