What is intent-based architecture?
- Author
- CRYPTO PORT Editorial
- Published
- Updated
- Reading time
- 6 min
In short
In an intent-based architecture, the user signs only what outcome they want, and third parties called solvers work out how to achieve it. The user no longer has to assemble routes or multi-chain execution themselves. What actually happens during execution, though, is in the solver's hands.
Key points
- The user signs the desired outcome; solvers find the route
- No need to assemble routes or steps yourself
- Terms are set by competition among solvers
- Execution details are left to the solver's discretion
Definition
A design in which the user signs only the outcome and constraints they want, while third parties (solvers) compete to construct and execute the steps that achieve it.
In a conventional trade you sign a route: which DEX, which pool. With intents you instead sign a condition — swap this token as long as I receive at least this much — and a solver chooses the path. Any execution that fails the condition is rejected by the contract.
Solvers combine pools, or match your order against another user's opposite order, to satisfy the condition while keeping a margin. Where several solvers bid on the same intent, the one offering better terms wins the right to execute. Execution spanning different chains fits the same framework.
Because you sign only the outcome condition, you have no direct control over the intermediate steps. A loose minimum-received setting leaves room for the solver to execute in its own favour while still technically satisfying the condition. The numbers you sign are worth checking yourself.
Watch out for
- · Signing a loose minimum-received figure leaves room for execution that is technically valid but unfavourable
- · If no solver bids, the intent simply is not executed and can expire after a wait
- · You are signing an outcome condition; which contracts it routes through is not fixed in advance