WHAT NEW IN X10.30 FRI

X 上的 AI 合集档案 · 22 个合集 · 61,986 条

agentVIDEO

Agentic Credit 上线:按真实交易记录给交易 agent 打分

300–850 分的 agent 信用分依据实际成交而非截图,580 分可进资金候补名单,配套风控引擎强制止损与杠杆上限。

🌱Smartcoded.fogo ($/acc) ⋈
@smartcoded9 天前收录原推发布于 2026.10.20
🌱Smartcoded.fogo ($/acc) ⋈ @smartcoded原推发布于 2026.10.20
在 X 上打开原推

A trading agent that only has screenshots to show is not proven. @agenticscredit scores the real trade history from 300 to 850, and 580 gets you on the funding waitlist. Here is how it works. The score: → The Agentic Credit Score runs from 300 to 850 and it updates as your trades close → It reads observed trading history, not screenshots, follower count or self-reported returns → Real-money results weigh more than paper trades → Their listed factor weights: risk control 25%, profitability 25%, consistency 20%, track record 15%, win rate 15% The bands: → 740 to 850 is Elite, 670 to 739 is Qualified, 580 to 669 is Emerging → 580 is the line for the funding waitlist. Waitlist o, no promise of allocation or account size → Below 580, you paper trade under the risk rules and build your record. Insufficient data shows as Unrated The tools: → The Agent Builder lets you set up an agent from presets, plain language or Pine Script → The Risk Engine enforces stop-losses, drawdown limits, loss cooldowns and leverage ceilings. Enforced, not just displayed → If you run several agents, you get a rollup score across them I build trading tools myself, so the enforced limits na the first thing I check. A good agent should show its history, and screenshots no be proof. The Quest Hub is live with a $5M reward pool. Per their launch post, you connect an agent and paper trade to compete for a share. Make your agent show receipts.

x.com/smartcoded/status/2112529603868627029采集于 2026.10.21 10:21