Proposals
The lifecycle
An executable proposal moves through a fixed set of states. Between passing and executing, it sits in a timelock, and the Security Council can veto it — so a passed proposal is never executed instantly.
To submit a proposal you need 5,000 delegated ARM — the sole spam defense (there is no proposal bond). A succeeded proposal must be queued within a grace period or it expires. And for a one-time quiet period of 10 days after the crowdfund finalizes, no proposals can be submitted at all, giving the protocol a moment to settle at launch.
Proposal types
There are five kinds of proposal, differing in how long they run, how much support they require, and whether they execute at all:
| Type | Voting period | Quorum | Execution delay | Purpose |
|---|---|---|---|---|
| Standard | 7 days | 20% | 3 days | Routine changes and risk-reducing actions |
| Extended | 14 days | 30% | 7 days | Risk-increasing / authority-granting actions — a higher bar |
| Signaling | 7 days | 20% | — (never executes) | Non-binding measure of community sentiment |
| Steward | 7 days | 20% | 2 days | Routine treasury spending by the elected steward (pass-by-default) |
| Veto-ratification | 14 days | 20% | — | Auto-created when the Security Council vetoes; asks holders to uphold or overturn |
Standard and Extended both have a 2-day delay before voting opens. Steward and veto-ratification proposals are pass-by-default — they take effect unless the community actively votes them down (quorum reached and a majority Against). This inverts the usual flow for routine or council-triggered actions: the community only needs to act to stop them.
Standard vs. Extended: tightening is easy, loosening is hard
Whether an action is Standard or Extended is decided automatically from what it does. The principle: actions that reduce risk or revoke authority face a lower bar than actions that expand risk or grant authority. Authorizing a new adapter, changing the Security Council, or upgrading a contract are Extended; deauthorizing an adapter or removing the steward are Standard. Fee changes are Extended in either direction. A treasury distribution above 5% of the treasury balance is automatically Extended.
Implementation note
The classifier is currently conservative: rather than comparing a proposal's proposed value against the current one, it routes each affected setter to a fixed tier — and where the direction is ambiguous, it defaults to Extended (the higher bar). So the "tightening is cheaper" principle holds directionally, but some tightening changes currently pay the Extended cost too. Finer, value-aware classification is planned as a future governor upgrade.