[2609 Grant] Kleros Foresight - Grant Completion Markets for RootstockCollective - Milestone 1

[JuanRah]

Before funding a grant, delegates assess whether a team can deliver. Kleros Foresight proposes adding a market-based estimate to that assessment: if this team is funded, will they complete their grant?

During a six-month pilot, participants will trade on grant outcomes, with completion defined by the terminal milestone payment. Delegates will receive a market reading before each funding vote, and the pilot will produce a public record of forecasts and outcomes to evaluate how useful those readings were.

Grant request: $15,000 in USDRIF, Strategic tier, across five milestones. This first request is $1,500 for Milestone 1. The money goes toward deploying the venue natively on Rootstock and running a six-month pilot. Kleros Cooperative covers the liquidity, the impermanent loss and the trading credits. The markets are non-binding and change no governance rules.

Sector: Analytics, dashboards & data infrastructure. Submission date: 21 September 2026.

Current stage: Reality.eth and the arbitrator are deployed on Rootstock mainnet. The Seer market contracts and the live interface are M2 deliverables; the Rootstock interface is currently a mockup.

Live mockup: https://rootstock2.vercel.app/

What Kleros delivers

A market that opens when a team posts its proposal and gives delegates a live estimate through the discussion window: the probability that this team, if funded, completes their grant.

Traders back their view with capital. If the DAO declines to fund, the market voids and positions are refunded. The DAO keeps every decision; the market makes completion risk visible earlier.

Markets will run on Kleros Foresight and settle via Reality.eth with Kleros as an arbitration backstop. They will trade in USDRIF on a native Rootstock deployment of Seer.

The output is decision-making data for the DAO and a public market venue on Rootstock. The pilot will test whether the market’s pre-funding estimates predict grant outcomes better than a historical baseline.

How the mechanism works

For each grant application we open a conditional market:

If {TEAM} is funded, will they successfully complete their grant?

• Market opens: When the off-chain proposal is posted to the forum.
• Voids and refunds: The application is withdrawn, outright rejected, or 120 days pass from the forum post with no first tranche paid. The void window closes permanently when the first tranche is paid.
• Resolves YES: The designated terminal milestone payment is made following its post-completion review.
• Resolves NO, after funding only: The DAO votes to cancel the grant, or 150 days elapse without a grant-tranche payment. The clock runs from the most recent grant-tranche payment and resets with each subsequent tranche.
• Pre-vote reading: The market’s published probability is recorded before the on-chain funding vote and published as the vote opens.

These are the proposed timing rules. M1 will finalize them with the Collective against grant schedules and terminal-review timing, before markets open.

FES remains the Collective’s framework for evaluating grant impact; it is not the market’s settlement source. Settlement uses the terminal payment, a timestamped on-chain event. KYC/KYB, AMAs, milestone reviews and monthly reporting stay as they are today.

Demand and relevant experience

The user of this grant is the Collective itself. This proposal has been through two rounds of written delegate review and has been circulated to the Collective’s working group. We thank Axia for BD context and review, and SEEDGov for substantive feedback on the mechanism. Delegates are invited to confirm in this thread whether they would use a pre-funding completion signal when assessing applications.

Kleros built and currently operates Kleros Foresight, with Seer providing the prediction-market infrastructure and Kleros handling dispute resolution. The team has experience running incentivised prediction-market experiments, including the Foresight movie experiment. Kleros team members also have direct operational experience with Deep Funding, which uses Seer infrastructure.

Liquidity and funding

Kleros Cooperative provides the liquidity and separately funds trading incentives for delegates and Rootstock community members. The liquidity allocation and credit budget will be confirmed during M1 before market launch; neither is requested from the Collective.

Credits are meant as working capital, not a payment: spendable only inside the markets, never withdrawable as cash. What a participant keeps is whatever their positions are worth at resolution. Participants may also trade their own capital alongside credits.

Cross-funding disclosure: No external grant funds the infrastructure this pilot runs on: Kleros funds the Foresight interface, the resolution layer and the pilot liquidity. For overlapping scope, the same Seer stack runs on Gnosis Chain, including a parallel decision-market pilot with Shutter DAO 0x36. Kleros offers comparable decision-market pilots to other DAOs as well. Kleros has received no funding from Rootstock Labs and no prior RootstockCollective grant.

Technical architecture and security

Markets use Seer contracts on Rootstock, collateralised in USDRIF, with the Kleros Foresight interface as the front end. Each market settles through Reality.eth on Rootstock, deployed at 0x4396DB62818eFd4dbcf34A069047E66dB4a1cc66. Disputes are decided by Kleros Court on Ethereum, with rulings relayed to Rootstock by a Kleros multisig.

The market contracts are not upgradeable. The arbitrator’s owner can set the Reality.eth address, set the dispute fee, publish arbitration metadata, and relay a Court ruling. It cannot make a ruling, only carry one. Ownership transfers to a 2-of-3 Kleros multisig before the first market opens, and the transfer transaction is part of the M2 evidence. The markets are advisory and cannot execute Collective treasury payments.

Prior security incidents disclosure: No security incident has involved the Kleros team, the Seer or Foresight protocol, or the Kleros Court contracts, on Rootstock or any other chain. The Reality.eth contracts this pilot deploys are audited, with no reported compromise. Reality.eth has featured in two third-party DAO incidents and in neither was any contract compromised or did any behave other than as specified. In both, an incorrect answer went unchallenged and was executed directly against a DAO treasury. Neither case was arbitrated by Kleros, and neither DAO was covered by Kleros Cooperative monitoring. Kleros Court is configured as the arbitrator from the first market and answer events are monitored, so an incorrect answer is challenged rather than left to finalize.

Milestones

M1 funds pilot design. M2 delivers the verified contracts, live Rootstock interface and end-to-end integration required before trading begins. M2’s delivery date will be established during M1, following engineering coordination, and published before the M2 funding request.

Milestone	Deliverables	Timing	Amount
M1 - Pilot design	Technical specifications: activation, void and resolution rules; pre-vote readings; credit distribution.	2 weeks	$1,500 USDRIF
M2 - Deployment	Verified Rootstock contract addresses, deployed versions and commits, live interface, USDRIF markets and resolution layer wired end to end.	TBC during M1	$3,750 USDRIF
M3 - Markets trading	Trading credits distributed, markets operating; markets analytics and reading report.	6 months of live markets	$3,000 USDRIF
M4 - Calibration and Final Report	Predicted versus actual outcome against the historical baseline; cost report, participation data.	4 weeks	$2,250 USDRIF
M5 - Terminal Milestone	Post-completion impact review against FES metrics	After M4	$4,500 USDRIF

Markets still open at the end of the pilot run to their own resolution. M4 reports on every market opened in the pilot, distinguishing settled outcomes from pending outcomes.

FES Metrics and reporting

Metric	Target/evidence
Market coverage	A market for every eligible grant application posted during the pilot window.
Readings before funding votes	A published reading for 100% of covered applications that reach an on-chain funding vote.
Participation	At least 15 delegate addresses issued credits and 10 distinct addresses trading over the six-month pilot.
Calibration	Every settled funded-grant market scored against its pre-vote forecast and a documented historical baseline.
Lasting infrastructure	Seer, Reality.eth and the arbitrator deployed, verified and public on Rootstock at pilot end.

Kleros will provide monthly activity reports on Discourse, milestone updates, and public links to markets and contracts. The final retrospective will detail trading volumes, active traders, liquidity, realized costs and participation. The calibration report is a research deliverable; outperformance against the baseline is not a condition promised by this proposal.

The infrastructure will remain available after the pilot. If targets are missed, Kleros keeps the venue operational past grant close and reports the shortfall and the fix. The proposal includes the program’s AMAs and 12-month post-grant communication obligations.

Team and resources

Juan Ramiro Garza (JuanRah) - Oracle Lead, Kleros. Leads mechanism design, resolution rules and coordination with the Collective. Previously Domain Allocator in the Education, Community Growth, and Events vertical in the Arbitrum DAO Grant Program. | GitHub | LinkedIn

Clément Lesaege - Co-founder and CTO, Kleros. Built Seer and owns the architecture of the market and resolution stack.

Alexandre (0xAlex) - Integration Lead, Kleros. Runs Kleros’s integrations and partnerships with DAOs/institutions.

Additional Links

Interface mockup | Kleros Foresight | Reality.eth repo | Reality.eth on Rootstock | Rootstock arbitrator | Kleros | Kleros (@Kleros_io) / X | Video pitch will be posted in this thread before the Initial AMA

[Raphael_Anode]

Very interesting proposal. In some ways it would also allow teams to hedge grants applications
But all in all I always loved decision markets for grants, and this is an implementation with human in the loop.

Thanks for proposing.

[Curia]

Thanks for proposing this, @JuanRah This reminds me of futarchy, although the prediction market here focuses on whether a project will complete its grant, providing an additional signal for delegates when evaluating grant proposals.

I’m curious whether you see the platform developing beyond this initial use case. For example, could it eventually support more detailed forecasting of grant-related outcomes, such as milestone delivery, adoption, or measurable ecosystem impact? Or is the intention to focus specifically on grant completion as the primary use case?

Do you see an opportunity to develop the platform into a prediction-market venue with wider applications beyond grant evaluation? If people could participate in forecasting different outcomes related to Rootstock’s ecosystem, this could open up new ways for the community to engage with governance and bring more attention to governance discussions.

I’d love to hear how you envision the platform evolving beyond the initial six-month pilot.

[Mattsanto]

My vote is IN FAVOR of Milestone 1 ($1,500 USDRIF) to move forward with the pilot design. My support for the upcoming stages will depend on how the team addresses the following key aspects:

1. Complexity for new users: If the goal is to get more members involved in the DAO, adding trading mechanics might overwhelm non-technical voters. How will you simplify the interface (UX) and what educational resources will you provide to prevent this from becoming a barrier to entry?

2. Manipulation risk: What mechanisms are planned to prevent a proposing team from massively buying the “YES” option with their own capital to artificially inflate their probability and bias the delegates?

3. Qualitative context behind positions: If the goal is to generate useful information for voters, is there a planned space where traders can leave arguments or reasons for their positions? A numerical probability is valuable, but knowing the underlying analysis enriches the debate and helps delegates in their own research.

[Seviramar]

I like the idea of adding a market-based signal to the grant evaluation process, especially because it creates a public record that can later be compared with actual outcomes.

One thing I’d be interested in understanding better is how you plan to account for markets with low participation or liquidity. For example, if only a few participants trade a market, would its probability still be presented to delegates in the same way as a market with much higher participation?

It could also be interesting to include some indication of market confidence or participation alongside the probability, so delegates have some context when interpreting the signal.

Overall, I think this could be a useful experiment, particularly because the pilot would generate actual data to evaluate whether the signal adds something to the existing grant evaluation process.

[SEEDGov]

Hi everyone!

Wanted to signal that we support this proposal. We believe this pilot program for futarchy and decision markets represents an interesting governance experiment that could benefit the Rootstock Collective in several ways.

First, it provides an opportunity to experiment with new approaches to decision-making, or, as in this case, with new tools designed to support the decision-making process. Second, from a visibility perspective, this experiment could position Rootstock Colkective at the forefront of experimentation with innovative governance tools, creating an opportunity to generate visibility, attract attention, and strengthen the DAO’s positioning within the broader governance ecosystem.

Finally, SEEDGov has previously promoted similar experiments with Kleros in other governance ecosystems, such as Velora, where unfortunately the protocol and DAO faced financial difficulties that forced the suspension of a program that was already underway and beginning to show promising results. We therefore see this initiative as a continuation of, and consistent with, the type of experimentation we have supported and pursued in the recent past, which is another reason why we are happy to support this proposal.

Lastly, in the interest of transparency, we want to disclose to the Collective that we have been in communication with Kleros since a very early stage of this proposal and provided feedback on the initial drafts. Those drafts were substantially different from the proposal currently under consideration, but we believe it is important to disclose our prior involvement nonetheless.

Thanks!

[IBUKI]

Thanks for sharing this proposal. I like the idea of adding a market-based signal to grant evaluation, especially because it gives delegates another data point before funding decisions without changing the governance process itself.

One thing I think will be important is how the probability is presented to users. A number like “72% chance of completion” can look more certain than it really is, especially if only a small number of traders are active or if liquidity is thin.

I would suggest showing the probability together with simple context such as the number of active traders, total liquidity, trading volume, and how concentrated the positions are. It would also help to clearly label the market as an advisory signal, not a recommendation or governance decision.

For the six-month pilot, I would also like to see whether the signal changes meaningfully as new information appears during the forum discussion. That could show whether the market is actually processing new information, rather than just reflecting the first impression of a proposal.

From a community perspective, a short guide explaining how to read the market and what its limitations are would also make participation easier for non-expert users.

Overall, I think the pilot could produce useful governance data, but transparency around market quality and interpretation will be important.