Autonomous On-Chain Supervision Agents for Cardano Treasury Projects, tackles a simple but critical question for Cardano: when the treasury funds a project, how do we know it is using money well and delivering what it promised?

Today, the answer is still mostly manual review. A small group of governance participants try to follow hundreds of teams and tens of millions of ADA. Reports are read by hand, evidence is checked one project at a time, and approaches differ between reviewers. From the outside, it is hard to see what was checked, when, and why a decision was made. This does not scale and it slows payments.

First, our solution puts clear milestones and rules directly on the Cardano blockchain. Each funded project gets concrete objectives and simple, checkable conditions written into smart contracts. These contracts hold funds in escrow and release the next payment only when the conditions are met. Every step leaves a permanent, public trail.

Next, we add AI-powered supervision agents. These software agents continuously watch public signals from funded projects, such as GitHub activity, documentation updates, and community channels. They compare what they see with the rules agreed in advance. When the evidence is strong that a milestone is complete, they trigger the transaction that releases funds. When something looks wrong or evidence is missing, they flag it for human review rather than paying out by default.

In practical terms, this creates a reusable supervision layer for all Cardano treasury programs. Treasury and governance participants can oversee many more projects with far less manual work. Builders spend less time writing long reports, because their real activity speaks for them. The wider community can open a simple dashboard that shows the status of projects, what was checked, and why each decision was made.

What makes this approach different is the combination of on-chain governance and off-chain AI. We move from occasional, subjective reviews to continuous, rules based, and transparent supervision. The system does not replace human judgement. It supports it with clear data, standard policies, and an auditable record that anyone can verify.

Who benefits from this? Treasury and governance actors gain objective information and a way to supervise hundreds of projects without a linear increase in effort. Funded teams get faster, more predictable payments when they deliver and keep their work open and visible. The broader Cardano community sees less “fund and forget”, more visible follow up, and better use of treasury funds.

The expected impact is significant. In the funded year, we aim to monitor at least twenty projects on mainnet, with between five and ten million ADA under active supervision. At least eighty five percent of their milestone payments should be released automatically based on clear evidence, and governance participants should see around a fifty percent reduction in manual verification time. Over the following years, the same system can scale to one hundred and fifty projects and around fifty million ADA per year, allowing Cardano to supervise roughly ten times more projects with the same governance capacity.

We are confident we can deliver this because of the team behind it. Our lead developer is an experienced Plutus smart contract engineer with successful previous Catalyst deliveries. Our AI specialist has years of experience in machine learning and anomaly detection on complex data. Our backend, infrastructure, and UI developers have run Cardano nodes, built dashboards, and shipped production systems. A dedicated project manager keeps us aligned with Catalyst processes, and specialist advisors support governance and security, including an external smart contract audit.

The total funding requested is 850,000 ADA over twelve months. This covers five milestones, from core smart contracts and rule engine to AI agents, data connectors, dashboards, full mainnet deployment with external audit, and governance handover. At scale, the effective cost is around 5,700 ADA per supervised project and about 17,000 ADA per one million ADA of treasury funds under protection. If this system helps the community avoid even a small percentage of under delivery on fifty million ADA of funding, it will have paid for itself several times over.

In conclusion, Autonomous On-Chain Supervision Agents for Cardano Treasury Projects turns today’s fragile, manual milestone checking into transparent, scalable, objective infrastructure. It helps the community supervise many more projects with the same effort, strengthens trust in treasury funding, and lays a foundation that future funds and programmes can reuse. With your support, we can make automated, verifiable project delivery a standard part of Cardano governance and protect the long-term health of the ecosystem. If you like this proposal, please vote for it in Cardano Catalyst Fund 15.

Thank you, we appreciate your help and support.