Compare
Silverscript vs Argent
Not rivals — a layer and the layer above it. The real question is whether your app is one contract or several that must talk.
| Silverscript | Argent | |
|---|---|---|
| What it is | A covenant language: one contract, its state, its rules | An app language on top: several actors (covenants) wired together |
| Unit of thought | The contract — old state + args + tx facts → allowed new state | The actor — entries that end in `become`, talking via ICC |
| Composability | N:M within one contract; covenant-id ownership | Actors + ICC — atomic cross-contract interaction |
| Compiles to | Kaspa Script directly | Silverscript, then Kaspa Script |
| Release state | v1.0.0 tagged 2026-09-09 — no external audit named | No tagged release; self-described "not yet release-ready" |
| Pinned dependency | — | Pins Silverscript to a pre-1.0 revision, not the release |
| Tooling | silverc compiler + CLI debugger; kascov.io playground | argentc build/inspect; examples in-repo (tickets, stones, ICC patterns) |
| Right for | Vaults, escrow, tokens, single-actor state machines | Multi-contract apps: ticketing, games, marketplace flows |
| Wrong for | Cross-contract apps wired by hand | Production funds today; anything one contract can do alone |
Pick Silverscript when…
The whole design fits in one contract's state machine: a vault, an escrow, a token, a lock. That's most covenant projects today, and it's the layer with a tagged release, a tutorial, a debugger, and the KCC20 book behind it.
Silverscript in full →Pick Argent when…
The design is genuinely several contracts interacting atomically — tickets and their venue, players and a game, a token and its minter. Prototype there today with the in-repo examples; treat production as blocked until it ships a release (checked 2026-09-10).
Argent in full →Facts on this page were checked against both repositories on 2026-09-10; release states also update daily on the status board.