​
Strategy Composer

Describe a strategy in one sentence — the Composer builds it as connected visual blocks on a canvas: triggers, swaps, and guardrails you can see and reason about. Every block is editable: open it, change an amount, a threshold, a schedule. Nothing runs blind: every strategy is simulated before any execution, so you see what would happen before anything does. Once live, a strategy runs 24/7 in its own dedicated wallet, watching the market even while you sleep. And you stay in charge: you sign and validate — the Composer never executes without your control.

​
Example: Buy the Dip

“When HYPE drops −5% from its recent high, buy 50 USDC of HYPE.”
That single sentence becomes a working strategy:
  • The trigger is checked continuously against the market — no manual chart-watching.
  • When HYPE falls 5% below its recent high, the strategy buys 50 USDC worth of HYPE — one buy per dip, never a runaway loop.
  • The trigger re-arms on every new high: HYPE recovers, sets a fresh high, and the strategy is ready for the next dip.

​
Strategy Templates

The Composer ships with a catalog of ready-made strategies for HyperEVM. Pick one, inspect the blocks, adjust, and run: Every template ships with guardrails baked in — minimum TVL, verified contracts, health-factor floors, peg-drift checks — visible as blocks on the canvas like everything else.

​
Available Blocks

Strategies are assembled from a small set of blocks that read like plain sentences:
  • Trigger / WHEN — “when HYPE drops 5% from its high”, “when APY beats my current vault”
  • Swap — “buy 50 USDC of HYPE”, “sell half back to USDC”
  • Stake — “stake HYPE for kHYPE”
  • Lend / Borrow — “borrow USDC against kHYPE”, “lend USDC at the best rate”
  • Guard — “keep my health factor above 1.5, unwind if it drops”
  • Repeat / Cron — “re-check every 6 hours”, “run once a day at 8:00”
Connect them on the canvas, and the flow reads top to bottom like the sentence you started from.

​
Working with Operators

A strategy doesn’t have to run alone. Hand it to an Operator and the agent takes over continuous execution: it monitors the strategy, reacts to its triggers, and queues any proposed action in an approval queue for you to review. The review-to-sign rule from the rest of the app applies unchanged — the Operator proposes, you approve.

​
Safety Model

  • One dedicated wallet per strategy. Each strategy runs in its own wallet, so its funds and permissions are isolated from your main wallet and from other strategies.
  • Preview before signature. Every execution is simulated and shown to you — exact amounts, routes, and outcomes — before your wallet is ever asked to sign.
  • Automatic unwind when risk rises. Guard blocks watch strategy health (like the lending health factor) and unwind positions automatically if risk crosses your threshold.