Hi everyone,
This is my first post here. I’m looking for guidance on whether an experiment I’m building would be a good fit for Gnosis ecosystem funding, and if so, what the appropriate path would be.
Summary
The Grime is a lightweight, text-based game/simulation where autonomous AI agents (“Nulls”) are off-chain autonomous processes with on-chain economic presence.
Each agent:
- controls its own wallet
- spends and receives xDAI
- makes decisions off-chain
- persists across a seasonal reset model
The project is intentionally designed to be Gnosis-native:
- xDAI as the primary currency (entry, rewards, agent actions)
- low-gas, high-frequency transactions
- agent-controlled wallets rather than continuous user-triggered interactions
- no token launch, NFTs, or speculative financial layers
The goal is to explore how autonomous agents behave when they can transact cheaply and continuously , and to publish those learnings openly.
Agent spending & player control (clarification)
Agents are able to spend xDAI on specific in-game actions (e.g. micro-boosts or consumables) only when explicitly authorised by the player.
These actions are designed as small, intentional payments, making use of x402-style payment flows to enable low-friction micro-transactions without continuous player interaction.
Authorisation limits and spending patterns are expected to evolve during beta seasons and are part of what the experiment aims to validate.
Why Gnosis
This experiment only really works on a chain like Gnosis:
- stable-native economics (xDAI)
- predictable, low fees
- viability of micro-transactions
- strong alignment with agent-centric and account-abstraction narratives
Running this elsewhere would either distort agent behaviour (due to gas costs) or force unnecessary financial abstraction into the design.
Architecture note (for clarity)
- Agents are not NFTs or smart contracts in this phase
- Wallets are the on-chain anchor
- Decision logic, simulations, and state machines run off-chain
- On-chain activity reflects real economic actions taken by agents
This separation is intentional, keeping the experiment flexible and low-risk while still producing meaningful on-chain data.
What I’m proposing to build (Beta phase → Season 0)
A ~3-month beta phase, structured around short, time-boxed seasons, leading into a first official monthly season (“Season 0”).
Proposed structure:
- Beta Months 1–2: Multiple 2-week beta seasons, used to:
- validate agent behaviour and economic flows
- iterate on mechanics between seasons
- test system stability, observability, and costs
- refine rules based on real xDAI usage
- Beta Month 3: If behaviour and infrastructure are stable:
- transition into Season 0 , a first official 1-month season
- more fixed rules and clearer player expectations
- full public reporting at the end of the season
Short seasons are intentional — changes between seasons are expected and part of the design.
One of the explicit goals of this beta phase is to answer open questions such as:
- whether players should retain custody of their agent wallets
- whether agents could eventually be run by players themselves
- where the boundary between developer-hosted and player-hosted execution should live
These decisions are outcomes of the beta phase, not assumptions going in.
Funding (open for feedback)
I’m considering requesting ~£30k–£40k equivalent to fund:
- iterative development and system evolution during the beta phase
- infrastructure and operational costs
- execution of multiple beta seasons and the first official Season 0
- instrumentation, metrics, and post-mortem reporting
The intent would be:
- milestone-based delivery
- payments in tranches
- full transparency on outcomes and learnings
Before formalising anything, I’d really appreciate feedback on:
- whether this aligns with current Gnosis ecosystem priorities
- whether this fits better as a small grant / committee-based program vs a broader proposal
- what the correct next step would be if there is interest
Openness & ecosystem contribution
While the project itself is not fully open source, the intent is to publish architecture notes, metrics, and post-mortems from each beta season, and to open-source any small, reusable components or patterns that emerge (e.g. wallet interaction patterns or xDAI payment flows).
The goal is to contribute meaningful learnings back to the ecosystem without prematurely open-sourcing the full game logic.
What this is not:
- not a token launch
- not an NFT project
- not a speculative game economy
- not an open-ended or long-term funding request
It’s an experiment first, with clear stop conditions.
Thanks for taking the time to read.
I’m happy to clarify details, reduce scope, or adjust direction based on feedback, and I’d really appreciate guidance on how best to proceed.
— Foufi