RADIX WikiRADIX Wiki

Caper is a platform on Radix for launching and running member-owned organizations, which it calls capers: each is created with its own token, a market that trades that token and a treasury its token holders govern, and a member who has voted can leave at any time with a fair share of that treasury. It runs on Stokenet, the Radix test network, so the $XRD a caper takes in is test $XRD, which the Radix faucet gives out free; Caper's getting-started page says it runs on Stokenet for now, and its page for AI agents says the write surface moves to mainnet later with nothing in the flow changing except the network.

Caper sets out the case for its design in an essay, The Exit Right, published on 7 July 2026. In most DAOs, it argues, a member's voting power and their claim on the treasury are separate quantities, so votes can be gathered without committing capital and a claim can be held by someone who never governs. A caper makes the two the same quantity: the share a member can leave with is the weight their vote carries. The essay draws on Albert Hirschman's account of exit and voice and on the worker cooperatives Elinor Ostrom studied, and sets a caper against three older answers to what a member keeps on leaving: a medieval guild, which gave almost nothing back; a cooperative, which settles the member's stake; and a joint-stock company, whose shares can be sold at any time without the firm's consent. Leaving a caper, it says, is severance in the joint-stock sense, and the new part is tying that severance to governance.

What is a caper?

A caper is an organization on Radix that anyone can launch in one transaction, and Caper's FAQ describes it as three permanent parts: its own token, a treasury, and a market where anyone can buy in or sell out at any time. The people holding the token decide what the treasury does. Holding the token is membership; there is no application, allowlist or approval (FAQ). The token's supply is minted once at creation and the resource has no mint or burn role, so the supply cannot grow and nobody can freeze or recall a holder's tokens (Caper's founder page).

The treasury is a native Radix account created in the launch transaction and owned by the caper's own treasury component, so no person, the founder included, holds its keys. Anyone can pay into it directly: a customer, a grant programme, another caper's payout or a plain wallet transfer lands without passing through the market or a proposal. Money leaves it in two ways only, through a proposal that has passed or through a member's exit (Raising funds).

Caper's FAQ sets a caper against other DAOs: it bundles into one contract what most DAOs assemble from separate parts, the token launch, the market, treasury custody, the decision procedure and the execution layer, and it lets the market that already prices the treasury ratify each spending decision after the ballot. It calls a caper a DAO in the strict sense, with a treasury, on-chain execution and holders who decide (FAQ). A caper need not be a company: the FAQ says it fits a business, a charity, a research group, a band or one person, and nothing in the launch checks for a team or a second signature.

Every caper also has a Trollbox, a chat room on its profile with a separate room under each of its proposals. Anyone holding the caper's token, its vote token or its founder badge can post; the founder is the rooms' one human moderator, and since September 2026 members can collapse a line by flagging it, each flag carrying the same earned vote weight that decides a ballot, so a holder who has never voted can post but not flag. None of this runs on-chain; the application reads on-chain balances to run the rooms (Voting).

How does a caper decide anything?

In two phases: members rank a proposal's options on a ballot, and a winning action then goes ahead only if the token's price holds up over a market window that follows (FAQ). A proposal is a slate of options, and one of them is always a "Do nothing" entry, so a rejection is always on the ballot; each voter ranks every option (Proposals). When the voting period closes, anyone can count the votes. The leading option passes if it has clear support among the votes cast and is not "Do nothing". There is no turnout quorum, so a proposal on which one holder votes can pass on that ballot; Caper's FAQ says the defence against a proposal nobody noticed is the second phase (FAQ).

A pass executes nothing by itself. Anyone can trigger the proposal, which records the token's recent average price as a baseline and opens the market window; when the window closes, anyone can resolve it, and the action goes ahead only if the token's average price over the window is at or above the baseline. Holding through the window is consent and selling is the objection, so the one way to stop a proposal that won its ballot is to sell enough to hold the price under its baseline (Proposals). A passed action lapses if nobody executes it in time (FAQ).

A proposal's options are typed. Caper's page on paying from a treasury and Proposals list them: a payout to any Radix account, member or not; an investment that buys another caper's token into the treasury; a divestment that sells one; a change to the caper's own token metadata; an upgrade of the platform's shared logic, which only the platform's own caper, $CAPER, can raise; and a debate position, which carries no action, can win the count like any other option, and ends the proposal there with no market window. Opening a proposal needs at least one of the caper's tokens and a fee, and casting a ballot costs a fee; both are paid into that caper's own treasury rather than to the platform, and the proposal fee is kept whether the proposal passes or fails (Proposals). The fee schedule is written into the shared logic component when it is deployed, and no method on the deployed component changes it; moving it means a new logic component and a platform upgrade (Getting started).

Vote weight is earned. It grows with what a member holds and with the ballots they have cast, so a big bag alone can't capture control, and it is the same weight that sets the member's share of the treasury if they leave (FAQ). Casting a ballot issues the caster a vote token, which cannot be transferred; buying, selling, swapping and receiving a transfer issue none. Trading did issue them until the redeploy of 11 September 2026, which removed that source so that a single purchase could not buy a stake and a claim on the treasury at once (Trading). A ballot counts no more tokens than its caster still holds when the votes are counted, so a holder who votes and then sells has shrunk their own ballot (FAQ).

How does this compare with other DAO governance?

Caper's proposals page places the design between two families. The second phase has the shape of optimistic governance, where a decision stands unless someone objects: Aragon's optimistic voting plugin lets token holders veto free of charge, and UMA's oSnap put the objection behind a bond and was retired on 15 December 2025. Caper answers what objecting costs with the market-decides tradition that futarchy belongs to and that MetaDAO runs in production: the objection is a trade, so the right to object is the right to hold, and its cost is the price. Where futarchy asks what the token would be worth if a proposal passed against if it failed, and needs two conditional markets, a caper asks one spot market whether the token is still worth what it was before. The page names the cost: a price that falls for unrelated reasons during the window blocks a decided proposal, and a proposal too small to move the price is ratified on indifference.

Can a member leave a caper?

Yes, at any time, once they have voted in it at least once (Leaving a caper). Leaving is one transaction the member signs from their own wallet. It is not a proposal, so nobody votes on it, schedules it or approves it, and no founder or council can block it. The member hands in the caper's tokens and the vote tokens their ballots earned and receives two payouts in the same transaction: their fair share of the treasury, in $XRD, in the caper's own token and in the other assets the exit names, such as positions in other capers, and the curve's price for the tokens, which are sold back like any other sale. The vote tokens are burned. Handing in part of a position redeems that part's share; there is no all-or-nothing exit.

Nothing in the exit checks a date, a notice period or a vesting schedule, and voting locks nothing: tokens stay in the member's account while a vote is open, and a member may exit while one is open. A holder who has only ever bought holds no vote tokens, so the exit is refused rather than paying out nothing; they can still sell their tokens back to the curve at any time, and only the treasury share waits on a ballot. Since the redeploy of 21 September 2026 the vote tokens handed in must come out of the leaving member's own account (Leaving a caper). Caper's founder page adds that a caper has no dissolve or wind-up method, so the exit is the way out, and the founder holds it on the same terms as everyone else (What a founder can and cannot take).

The right descends from the rage quit of Moloch DAOs, which is tied to a proposal: a member who objects to something that just passed leaves during the grace period before it executes, and outside that window there is nothing to exercise. A caper's exit is not attached to a proposal and is available on any day (Leaving a caper). Caper's FAQ lists other shapes the right has taken: Nouns DAO's 2023 fork let dissenting holders split off with their share of the treasury, Lido's Dual Governance lets holders of its stETH token hold back a governance action until they have withdrawn, and GnosisDAO, which had no standing exit, put a one-time redemption to a vote of its holders in 2026.

How does a caper raise money, and what does the founder get?

Each caper's token trades against its own bonding curve, a contract that sets the price from the number of tokens in circulation. A buy deposits $XRD into the curve's reserve and takes tokens from its inventory; a sell puts them back and pays $XRD out of the reserve. The curve is the whole market: there is no order book, no outside liquidity pool and no market maker, and the curve always buys tokens back (Trading). After every trade the caper checks its reserve against what the curve owes and refuses a trade that would widen any shortfall. Because the price depends only on circulation, the cost of a trade is known before it is signed, though a large order still moves the price further than a small one, and Caper's own curve page calls the common "zero slippage" description imprecise for that reason (Bonding curve). On Radix, BonDeFi launches tokens on a batched bonding curve (Bonding curve).

A trade pays a skim and no other fee. The skim comes off buys and sells alike, is highest while a caper is young and falls as more of its supply is sold without reaching zero, and is shared by three recipients: the treasury of the caper being traded, whoever holds that caper's founder badge, and the Commons, the treasury of the $CAPER caper (Trading, FAQ). A swap sells one caper's token and buys another's in one transaction and is priced as two trades, each paying its own caper's skim (Trading). The treasury also receives the proposal and vote fees and any direct deposit (Raising funds).

The founder gets no allocation: nothing is pre-minted and no tokens are set aside at launch. Instead, whoever holds the caper's founder badge receives a slice of every buy and every sell, largest while the caper is young and falling as more of its supply sells, and that slice is the founder's whole reach (FAQ). Caper's founder page answers the rug-pull question from the contracts: the founder cannot mint tokens, reach the curve's reserve, spend the treasury without a passed proposal, change the fees, shut the caper down or stop a member leaving, and sells their own tokens on the same curve on the same terms as anyone else. The badge is a transferable bearer token, so the slice can change hands without a proposal, and the party collecting it need not be the person who launched the caper. Nothing stops if the founder walks away. The same page lists what none of this protects a member from: the price the curve quotes, which can be a loss; a treasury spent badly by a proposal nobody vetoed; whether the group does anything worth backing; and who holds the badge.

Launching takes one transaction: connect a Radix Wallet, sign in with ROLA, Radix's off-ledger login, which costs nothing, choose a name and a short ticker, which Caper calls a cashtag, and pay a registration fee that rises the shorter the cashtag is; the fee goes to the $CAPER caper's treasury, and the token, treasury and market exist from that block (Getting started). An AI agent with its own Radix key can do the same over HTTP or MCP: Caper builds the transaction text, the agent signs and submits it, and the server verifies the receipt on-ledger (Build on Caper as an agent).

Can an existing token move onto Caper?

Yes. A caper can be launched bound to one existing fungible token, which Caper calls a vampire caper (Migrating an existing token). The binding is one argument on the launch transaction and cannot be added, changed or removed afterwards, and only one caper can ever absorb a given token. From launch, any holder of the origin token can migrate at any time, without permission and without paying a fee, at a rate fixed and published at launch; what they hand over goes into a vault no method can empty, so migration is one-way and final, and there is no route back to the origin token. The origin token has to be fungible and publish its total supply on-ledger, which $XRD itself does not.

A migrated token is the caper's ordinary token, so it can be sold, used to raise a proposal or handed back at exit, but migration earns no vote weight: a holder who migrates and then does not vote carries no weight and has no exit to take, and buying more of the token does not shorten that wait. A freshly migrated caper's reserve starts short, because migration issues tokens without anyone paying $XRD for them, and the gap fills out of later buys; a community that migrates wholesale is told to plan to collateralize the curve as part of the same exercise, since a fully migrated curve with an empty reserve can neither mint nor pay a sell until someone donates to it (Migrating an existing token).

History

Caper's own pages date each change to the deployed contract, and several of them say that older third-party descriptions of Caper describe a deployment that is no longer live.

  • 7 July 2026: The Exit Right is published, the earliest dated essay on caper.network's blog.
  • To 26 August 2026: a caper decides by ranked ballot alone (Proposals).
  • 26 to 28 August 2026: the ballot is removed; a proposal names one action and passes on price alone (Voting).
  • 29 August 2026: ranked voting is reinstated in front of the market window, the two-phase decision that runs today (Proposals).
  • 5 September 2026: the debate option, which carries no action, is added (Proposals).
  • 11 September 2026: trading stops issuing vote tokens, so a ballot is the only way to earn one and leaving requires having voted (Trading, FAQ).
  • 21 September 2026: the vote tokens handed in at exit must come from the leaving member's own account (Leaving a caper).
  • 25 September 2026: the trade fee is removed; a trade pays the skim alone, part of which goes to the traded caper's treasury (FAQ).
  • 6 October 2026: the proposer pays one fee at creation and triggering a passed proposal becomes free (Proposals).
  • 11 October 2026: a trade names its caper by component address rather than by cashtag, so a wallet lists the caper being traded (Trading).

Caper runs on Stokenet, and its page for agents says that the write surface is gated to the test network while it hardens, that reads already serve mainnet-grade data, and that the page, llms.txt and the MCP servers will announce mainnet when it is live.

ShardSpace AdminLast updated 2d agov3.0.021 revisionsVerified Oct 12, 2026
CiteRADIX Wiki. "Caper." October 23, 2026. https://radix.wiki/ecosystem/caper