WatchDog

[EmmanuelEklipse]

Project Abstract

WatchDog is a unified grant tracking platform purpose-built for the Web3 Foundation. It aggregates updates from GitHub—pull requests, issues, commits—and other public channels into a streamlined, reviewer-friendly dashboard, enhancing transparency and accountability across W3F-funded projects. By leveraging machine learning to monitor milestones, detect delays or inactivity, and support on-chain progress attestations, WatchDog reduces manual oversight, highlights project impact, and enables more scalable and effective grant management for successful grant outcomes.

If your application is a follow-up to a previous grant, please mention which one in the first line of the abstract and include a link to previous pull requests if applicable.

Grant level

• Level 1: Up to $10,000, 2 approvals
• Level 2: Up to $30,000, 3 approvals
• Level 3: Unlimited, 5 approvals (for >$100k: Web3 Foundation Council approval)

Application Checklist

• The application template has been copied and aptly renamed (project_name.md).
• I have read the application guidelines.
• Payment details have been provided (Polkadot AssetHub (USDC & DOT) address in the application and bank details via email, if applicable).
• I understand that an agreed upon percentage of each milestone will be paid in vested DOT, to the Polkadot address listed in the application.
• I am aware that, in order to receive a grant, I (and the entity I represent) have to successfully complete a KYC/KYB check.
• The software delivered for this grant will be released under an open-source license specified in the application.
• The initial PR contains only one commit (squash and force-push if needed).
• The grant will only be announced once the first milestone has been accepted (see the announcement guidelines).
• I prefer the discussion of this application to take place in a private Element/Matrix channel. My username is: @_______:matrix.org (change the homeserver if you use a different one)

[github-actions]

CLA Assistant Lite bot All contributors have signed the CLA ✍️ ✅

[EmmanuelEklipse]

I have read and hereby sign the Contributor License Agreement.

[keeganquigley]

By "grant recipients" are you referring to child bounties here? Or just recipients of treasury proposals in general?

[EmmanuelEklipse]

Thanks for taking the time to review this, Keegan. By "grant recipients, we refer to all recipients of Polkadot Treasury funding, including both direct Treasury proposals and child bounties. Although the main priority is focused on helping W3F grant oversight. WatchDog is designed to provide comprehensive oversight for all funded initiatives within the Polkadot ecosystem.

[keeganquigley]

Thanks @EmmanuelEklipse I ask because the W3F grants program is completely separate from the treasury and OpenGov at this time, so I think to keep this proposal open I would ask you to take out the treasury parts and focus solely on W3F grants.

[EmmanuelEklipse]

Roger that. Done

[keeganquigley]

So $10k for the integration but then how much for maintenance costs moving forward?

[EmmanuelEklipse]

The $10k requested covers the development and deployment of the WatchDog MVP (Milestones 1 and 2). For ongoing maintenance, we estimate annual costs of approximately $300/month, covering server hosting, API upkeep, data sync pipelines, and minor updates. This keeps the MVP running. If additional features or integrations are needed beyond this scope, we’ll submit a separate proposal so the W3F can evaluate based on value delivered.

[keeganquigley]

If this dashboard is mainly geared towards OpenGov members in my opinion you should ask for treasury funding first since they will be the primary users.

[EmmanuelEklipse]

WatchDog is built as a public good for the entire Polkadot ecosystem, serving the Web3 Foundation, OpenGov participants, DOT token holders, grant recipients, and the broader community. While it offers valuable insights for governance and voting within OpenGov, its primary goal is to address the Web3 Foundation’s operational challenges around grant oversight.

We’re requesting funding from the Web3 Foundation because WatchDog’s core capabilities, such as on-chain milestone verification using hashed attestations and Substrate-based data synchronization, are tightly aligned with W3F’s mission to strengthen the infrastructure of Polkadot’s decentralized ecosystem.

• The platform reduces W3F’s administrative burden by automating time-consuming manual tasks. For example:
• It automatically follows up with grantees who miss deadlines, eliminating the need for manual check-ins.
• Its machine learning system monitors real-time data (e.g., GitHub activity, grant metadata) and flags risks such as delays, inactivity, or unusual patterns. When risks are detected, WatchDog sends automated alerts to administrators, enabling proactive intervention.
• It will also make it easier to vote on grants by admins as showcased in our live demo (https://watchdog2.vercel.app/alt-view),
  

WatchDog delivers easier visualizations and alerts, making it easy for administrators to identify pending approvals and review key performance metrics at a glance. It also makes it easier for admins to track the impact after the grant is completed.
As the Polkadot ecosystem expands, W3F needs scalable tools to manage an increasing number of grants and data streams without compromising the quality of oversight. By focusing the MVP on W3F’s immediate needs, WatchDog sets a strong foundation for broader adoption across the ecosystem.

[keeganquigley]

Something like this could possibly be helpful for us, but it would need to be limited in scope to the W3F grants programs, in order for us to fund it. Since our programs are separate buckets as I mentioned, we currently ask for treasury related proposals to be funded by the treasury itself. With this option it might also be easier to maintain since our regular grants program is entirely maintained on GH.

[EmmanuelEklipse]

You are right, I have made the changes. Thanks.

[keeganquigley]

Thanks for the changes @EmmanuelEklipse much appreciated. I will mark the application as ready for review and ping the rest of the committee for comment.

[EmmanuelEklipse]

@keeganquigley Thanks, would this be a tool that doesn’t add value to the W3F?

[EmmanuelEklipse]

We are planing to an auto review feature to provide initial feedback for applicants in regards to missing details in proposals.

[keeganquigley]

Thanks @EmmanuelEklipse could you also add some initial mockups for the frontend UI? We usually ask for these upfront, even if they are rough.

Additionally, all the links to the existing pictures in your application are broken now. Can you fix these? Thanks!

[semuelle]

Thank you for the application, @EmmanuelEklipse. I like the mockups, and it seems you already put a fair bit of thought into the platform. That being said, we have already experimented with web-based grants management platforms (https://grants.web3.foundation/docs/RFPs/grant_management_webapp), but found the process of parsing Markdown documents and pull requests too error-prone.

I recommend that you start a conversation with the community to investigate whether this could be used by bounties. You could post in the Polkadot Forum or reach out to some people who are already working on bounty tooling, such as the Bounty Manager team.

[EmmanuelEklipse]

@semuelle Thank you for the thoughtful feedback. I appreciate you recognizing the detail we've put into WatchDog, regardless of the proposal’s outcome.
You're right about the challenges with parsing Markdown and PRs. While AI has improved this area, our approach goes beyond parsing by combining structured grantee inputs, targeted GitHub signals, and lightweight ML to make milestone tracking more reliable and auditable.
How WatchDog Tackles the Parsing Challenge:

• Structured Tags: Grantees use standardized tags (e.g., #milestone-1-complete) in PRs/commits, providing clear machine-readable signals.
• Standard Template Updates: Use lightweight, consistent progress templates (YAML-style in Markdown) to reduce ambiguity and improve automation.
• AI as a Supplement, Not the Core: AI/NLP serves as a fallback, enhancing, not replacing, structured communication.

Why This Matters: We're not trying to infer intent from messy Markdown; we're creating a workflow that encourages clear signals and enhances reviewer efficiency, especially across multiple grants or bounties.

Thanks for the suggestion to start a conversation with the community to investigate whether this could be used by bounties. We will do that, as well as reach out to some people who are already working on bounty tooling. Our vision is to streamline operations and reduce manual oversight for the W3F and the broader DOT ecosystem. If this were a tool to be used by the W3F, we plan on introducing AI Grant Review to provide preliminary feedback for applicants regarding missing details and fit in proposals. Thank you again for your time and feedback.

[EmmanuelEklipse]

@keeganquigley @semuelle Thank you for the feedback. Closing this proposal.

[keeganquigley]

Closed per discussion in #2575 (comment)