Futures
Access hundreds of perpetual contracts
CFD
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Pre-IPOs
Unlock full access to global stock IPOs
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
Promotions
AI
Gate AI
Your all-in-one conversational AI partner
Gate AI Bot
Use Gate AI directly in your social App
GateClaw
Gate Blue Lobster, ready to go
Gate for AI Agent
AI infrastructure, Gate MCP, Skills, and CLI
Gate Skills Hub
10K+ Skills
From office tasks to trading, the all-in-one skill hub makes AI even more useful.
GateRouter
Smartly choose from 40+ AI models, with 0% extra fees
"Markdown fully disabled": Engineer Claude recommends having AI generate HTML directly
According to Beating Monitoring, Thariq, a member of the Claude Code team under Anthropic, said in a post that he has completely stopped using Markdown in his day-to-day work and instead lets AI generate HTML files by default. He believes that as AI-generated plans become increasingly complex, HTML’s advantages in information density and two-way interaction far outweigh those of plain text.
Thariq noted that Markdown longer than 100 lines is extremely hard to read and also forces large models to draw diagrams using garbled ASCII characters. After switching to HTML, not only can it present architecture diagrams and interface designs more intuitively with CSS and SVG, but it can also enable deeper interaction.
The most essential use case is to have the AI build temporary editing interfaces. When he encounters requirements that are difficult to describe in words, he asks Claude to create a temporary webpage with draggable cards or color sliders, along with a “Copy as JSON” button. After tuning the results in the webpage by instinct, he directly copies the parameters back into the command line to feed into the large model, completing the loop.
Although it takes the large model 2 to 4 times longer to write HTML than Markdown, and it’s even more difficult to compare differences (Diff) in version control, he said that this kind of experience—pulling people back into the decision-making cycle—is absolutely worth it.