On‑Chain Smart Contract Classifier Library (Python/JS) 📚

[Proposal setup] Proposal title

Please provide your proposal title

On‑Chain Smart Contract Classifier Library (Python/JS) 📚

[Proposal Summary] Budget Information

Enter the amount of funding you are requesting in ADA

45000

[Proposal Summary] Time

Please specify how many months you expect your project to last

8

[Proposal Summary] Translation Information

Please indicate if your proposal has been auto-translated

No

Original Language

en

[Proposal Summary] Problem Statement

What is the problem you want to solve?

Cardano lacks tools to classify deployed Plutus smart contracts by type or usage This makes it hard for developers, auditors,researchers to understand ecosystem dynamics or discover relevant contracts

[Proposal Summary] Supporting Documentation

Supporting links

• https://github.com/cardano-foundation/cf-gsoc-ideas-page-2025
,
• https://arxiv.org/abs/2503.09327

[Proposal Summary] Project Dependencies

Does your project have any dependencies on other organizations, technical or otherwise?

No

Describe any dependencies or write 'No dependencies'

No direct dependencies.Will use: Cardano-db-sync or Koios for blockchain data Python (scikit-learn), TensorFlow, Node.js CIP‑67, CIP‑68 for metadata standards

[Proposal Summary] Project Open Source

Will your project's outputs be fully open source?

Yes

License and Additional Information

Yes, entire project will be released under MIT License. Code and datasets will be available publicly via GitHub from month one.

[Theme Selection] Theme

Please choose the most relevant theme and tag related to the outcomes of your proposal

Developer Tools

[Campaign Category] Category Questions

Mention your open source license and describe your open source license rationale.

We will release the entire codebase under the MIT License, a permissive OSI-approved license that allows anyone to use, modify, distribute, and contribute to the project freely. We chose MIT to ensure maximum adoption and flexibility for both individual developers and enterprises, while maintaining proper credit to original authors. This approach encourages collaboration and removes commercial or legal barriers for reuse.

How do you make sure your source code is accessible to the public from project start, and people are informed?

We will create a public GitHub repository at the beginning of the project and commit all source code, datasets, and roadmap documents from day one. Updates will be pushed regularly, and the repository link will be shared on our Catalyst proposal page, Twitter, and Cardano community forums. Contributors and testers will be encouraged to join via README and contribution guidelines.

How will you provide high quality documentation?

We will maintain comprehensive documentation in a /docs folder and GitHub Wiki, including:

Installation & usage guides

CLI and API references

Architecture overview

Code comments and examples

Tutorials and video walkthroughs

We’ll also include multilingual documentation (starting with English & Amharic) to support global accessibility and contribution.

[Your Project and Solution] Solution

Please describe your proposed solution and how it addresses the problem

This tool will extract features from on-chain smart contracts and train ML models to classify contracts into categories like:

DeFi (DEXes, lending)

NFT minting

DAOs and governance

Games

Treasury & multisig tools

The output will be available via:

A CLI tool for developer use

A REST API for explorer integration

A web dashboard showing trends

[Your Project and Solution] Impact

Please define the positive impact your project will have on the wider Cardano community

Helps researchers and builders better understand smart contract patterns

Improves discoverability of useful contract templates

Assists auditors with anomaly detection

Enables developers to analyze the ecosystem easily

Useful for dashboards, explorers, and academic reports

[Your Project and Solution] Capabilities & Feasibility

What is your capability to deliver your project with high levels of trust and accountability? How do you intend to validate if your approach is feasible?

Team has experience with Cardano node data, Plutus, Python, and machine learning

MVP will use scikit-learn and Koios

Future versions may include deep learning and auto‑feature generation

Designed to be modular and community‑extendable

[Milestones] Project Milestones

Milestone Title

Repo Setup & Data Pipeline Initialization

Milestone Outputs

We will create the core structure of the project, including a public GitHub repository with a well-organized codebase and an automated pipeline to fetch and parse Cardano smart contract data. This phase will include integration with Koios or Cardano-db-sync to retrieve smart contract metadata, transaction history, and Plutus script hashes. We will also design a feature extraction module to process this data for use in training the classification model.

A developer-friendly CLI will be scaffolded with placeholder functionality for future integration. We will also publish the initial documentation: project vision, architecture plan, contribution guidelines, license, and how to run the initial pipeline locally.

We will announce the project in developer communities and open it up for early feedback and issue reporting.

Acceptance Criteria

GitHub repo is public and licensed under MIT

Project structure is modular, with folders for data, models, CLI, and docs

Koios (or db-sync) integration pulls and saves smart contract metadata

Feature extraction script outputs structured JSON

CLI includes placeholder structure

Initial README and contribution guide available

Announcement made on Catalyst and Twitter/X

Evidence of Completion

Live GitHub repository

JSON output samples (smart contract features)

Screenshots or demo video showing data fetch

CLI structure shown in terminal

Screenshot of public community announcement

Delivery Month

2

Cost

7000

Progress

10 %

Milestone Title

Model Training & Smart Contract Classification Engine

Milestone Outputs

In this milestone, we will develop and train the first working version of the smart contract classification model. Using the features extracted in Milestone #1, we will apply supervised machine learning techniques to label contracts into categories such as DeFi, NFT minting, treasury/governance, games, and more.

We will experiment with classification models using scikit-learn (Random Forest, Decision Trees, Logistic Regression) and validate results using cross-validation and confusion matrices. The classification logic will be integrated into the CLI, allowing users to analyze individual or batch smart contracts locally.

An internal evaluation dataset will be created, and documentation will explain the model’s assumptions, inputs, and output format. This version will support offline/local use for privacy and developer testing.

Acceptance Criteria

Trained model with at least 70% accuracy on test data

CLI supports basic classify command

Output includes category and confidence score

Model and training scripts pushed to GitHub

Evaluation dataset (open format) published

Documented classification process and how to replicate training

Classifier can run on local machine without cloud dependency

Evidence of Completion

GitHub repo with full model code and training logs

Sample outputs for several contract IDs

Screenshot of classification results via CLI

Evaluation report with accuracy, precision, recall

Public announcement and example post (e.g. “Classified 100 random smart contracts: results”)

Delivery Month

4

Cost

9000

Progress

20 %

Milestone Title

REST API & Public Web Dashboard (v1 Beta)

Milestone Outputs

This milestone focuses on expanding the accessibility of the classifier through a web-based REST API and a simple public dashboard interface. The API will allow users and developers to send contract addresses and receive classification results (category, confidence, metadata). The backend will be built using Node.js (or FastAPI for Python), securely hosting the model and processing user queries.

We will also build a basic web dashboard to allow anyone—without CLI knowledge—to input a contract hash and see live classification results. The dashboard will visualize trends such as category distribution, high-activity contract types, and historical data snapshots.

This web interface and API will be hosted on a reliable cloud platform (like Vercel, Render, or DigitalOcean) and will include rate limiting, API keys (optional), and CORS headers for frontend integrations.

All API documentation (endpoints, usage examples, error codes) will be included in the /docs section and the GitHub Wiki.

Acceptance Criteria

REST API live and accessible over public web

Supports POST or GET to classify a given contract ID

Web dashboard online, responsive on desktop/mobile

Endpoint documentation with examples published

At least 3 test contracts displayed in the dashboard

GitHub includes instructions for self-hosting API

API tested for speed, accuracy, and uptim

Evidence of Completion

Public URL of API and dashboard

Screenshots of classification interface and results

Link to docs with code examples and API usage

GitHub repo updated with server + dashboard source

Feedback form embedded in dashboard

Metrics showing uptime + successful classifications

Delivery Month

5

Cost

8000

Progress

10 %

Milestone Title

Community Beta Testing & Feedback Integration

Milestone Outputs

This milestone focuses on engaging the developer and Catalyst community to test the API, CLI, and dashboard and collect valuable feedback for refinement. We will launch a public beta test program where users can try the classifier on live contracts and provide suggestions or bug reports through GitHub issues, a simple feedback form, or Telegram/Discord group.

We’ll gather data on classification accuracy, UX friction, bugs, and improvement requests. Based on this feedback, we will refine the model, adjust categories, improve error handling, and polish the UI. We'll also document known limitations and use cases, adding a Q&A or FAQ page to guide users.

A community leaderboard or recognition system (GitHub contributors, testers) may be added to increase involvement. Multilingual outreach will be piloted with translations into Amharic and Arabic for documentation and UI.

Acceptance Criteria

Minimum 20 community testers engaged

At least 10 unique bug reports or suggestions received

Feedback from 3+ countries/language groups

UI/UX updates implemented based on feedback

Improved classification model based on error trends

FAQ and Known Issues documentation published

GitHub “Contributors” and “Testers” pages added

Evidence of Completion

Screenshots of user feedback and chat discussions

Public list of testers/contributors

Change log showing improvements from community input

Updated classifier accuracy metrics

Before/after screenshots of dashboard

Multilingual versions of doc or UI (Amharic, Arabic)

Delivery Month

6

Cost

6000

Progress

20 %

Milestone Title

Final Release v1.0 + Documentation & SDK Publishing

Milestone Outputs

In this milestone, we will finalize and release version 1.0 of the classifier library, REST API, CLI, and dashboard. This includes optimizing the codebase, testing edge cases, and polishing the UI/UX for production readiness. A major focus will be on shipping high-quality, developer-friendly documentation and SDKs in both Python and JavaScript, making integration into any project seamless.

We will publish full installation guides, usage examples, model performance benchmarks, data schemas, error handling info, and REST API specs. For CLI users, we’ll also include common use cases, CLI flags, and example commands. Each SDK will include NPM and PyPI packages with usage instructions and examples.

Additionally, we will record short video tutorials and walkthroughs and publish them on YouTube, along with full markdown guides in the GitHub Wiki.

This milestone will also include localization of documentation and UI for at least two languages beyond English, starting with Amharic and Arabic, supporting inclusive global adoption.

Acceptance Criteria

Classifier tool released as v1.0 on GitHub with version tag

Python and JS SDKs available on PyPI and NPM

Complete and clean documentation published (README, API, CLI, model)

GitHub Wiki populated with dev-focused resources

UI/CLI support multi-language label output

At least 2 full YouTube tutorial videos live

Final dashboard online with improved UX and performance

Multilingual support confirmed for docs and labels

Evidence of Completion

Public GitHub v1.0 release page

PyPI and NPM links for SDKs

Live tutorials on YouTube (linked from README)

Screenshots of final dashboard and CLI

Verified functionality of SDKs and example projects

Translated docs in /docs/lang/ folders

API uptime and performance metrics

Delivery Month

6

Cost

9000

Progress

20 %

Milestone Title

Final Reporting, Community Handoff & Ecosystem Integration

Milestone Outputs

In the final milestone, we will complete project reporting, provide handoff documentation for long-term community ownership, and begin outreach to ecosystem projects for adoption and integration. We'll submit a detailed final report to Project Catalyst, summarizing deliverables, usage statistics, feedback received, and lessons learned.

To ensure sustainability, we’ll also provide:

A full "Maintainer's Guide" to help new contributors continue development

An “Integration Guide” for explorers, dashboards, or analytics platforms

Outreach to Cardano developer teams (e.g. Aiken, Plutus, Koios, or community explorers) for potential API or SDK integration

We'll archive all source code and documentation on GitHub under versioned releases, with tagged commits and clear changelogs. A permanent live demo or mirror site may be hosted to showcase ongoing functionality. Future update plans will be outlined in a roadmap document for continued iteration.

We’ll also present results in a Twitter thread, Reddit post, and submit to relevant Cardano community newsletters.

Acceptance Criteria

Final report submitted to Catalyst with clear KPIs and deliverables

“Maintainer’s Guide” and “Integration Guide” uploaded to GitHub

Outreach emails/messages sent to at least 5 ecosystem projects

Final recorded video presentation or community livestream

Roadmap doc added to GitHub

Optional: live community AMA or showcase event (Discord/Twitter)

Evidence of Completion

Final report (PDF or MD format) shared publicly

Screenshot of GitHub guides and roadmap

Record of communication with partner projects (emails or DMs)

Screenshots or video of community engagement (AMA, Reddit, Twitter)

Evidence of GitHub versioning and repository completeness

Delivery Month

8

Cost

6000

Progress

20 %

[Final Pitch] Budget & Costs

Please provide a cost breakdown of the proposed work and resources

💰 Budget & Cost Breakdown (Total: 45,000 ADA)

| Category | Cost (ADA) | Description |

| -------------------------------- | ---------- | -------------------------------------------------------------------- |

| Development & Engineering | 25,000 | ML model creation, backend API, dashboard, CLI tooling |

| Documentation & Localization | 5,000 | Full technical documentation, multilingual support (Amharic, Arabic) |

| Community Testing & Feedback | 3,000 | Incentives for testers, feedback platform setup |

| Hosting, DevOps & Deployment | 2,000 | API + dashboard hosting (VPS), uptime monitoring |

| Final Packaging & SDK Publishing | 4,000 | SDKs for Python/JS on PyPI/NPM, release engineering |

| Project Management & Reporting | 6,000 | Catalyst progress updates, proposal reports, roadmap delivery |

| Total | 45,000 | — |

[Final Pitch] Value for Money

How does the cost of the project represent value for the Cardano ecosystem?

This project offers high value to the Cardano ecosystem by delivering:

✅ A first-of-its-kind classification library for Cardano smart contracts a powerful tool for explorers, researchers, and developers.

✅ Multiple access methods: REST API, CLI tool, SDKs, and web dashboard all free and open source under MIT license.

✅ Improved transparency and discoverability in the Cardano smart contract space enabling security researchers and analytics platforms.

✅ Built-in localization support helping grow adoption in non-English speaking regions (e.g. Ethiopia, MENA).

✅ Reusability and community ownership full code and models will be open for integration into other tools and maintained by contributors long-term.

Compared to traditional software development costs, 45,000 ADA is a lean budget that maximizes ecosystem growth, developer engagement, and decentralized collaboration.

[Required Acknowledgements] Consent & Confirmation

Terms and Conditions:

Yes