[ADApter] MCP Server and AI Agent for Cardano

[Proposal setup] Proposal title

Please provide your proposal title

[ADApter] MCP Server and AI Agent for Cardano

[Proposal Summary] Budget Information

Enter the amount of funding you are requesting in ADA

100000

[Proposal Summary] Time

Please specify how many months you expect your project to last

6

[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?

The world is shifting toward AI agent systems. Without infrastructure for them to access Cardano’s on-chain data, the blockchain risks limited interoperability and lagging utility.

[Proposal Summary] Project Dependencies

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

Yes

Describe any dependencies or write 'No dependencies'

Cardano’s On-chain: [Blockfrost.io, CardanoScan]; LLM: [Claude Sonnet]; Search & Crawling: [Firecrawl]; Market Information: [CoinMarketCap]; Catalyst Insights: [LidoNation];

[Proposal Summary] Project Open Source

Will your project's outputs be fully open source?

Yes

License and Additional Information

All outputs will be public under Apache 2.0 license

[Theme Selection] Theme

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

AI

[Campaign Category] Category Questions

Describe what makes your idea innovative compared to what has been previously funded (whether by you or others).

Our innovation is the creation of Cardano's first Model Context Protocol (MCP) server, a foundational infrastructure layer that allows any AI agent to natively access on-chain data which has not been funded before.

Describe what your prototype or MVP will demonstrate, and where it can be accessed.

Our MVP will demonstrate real-time querying of Cardano data, accessible to everyone through a public website and an X/Twitter bot. The entire project will be fully open-source and accessible on GitHub.

Describe realistic measures of success, ideally with on-chain metrics.

Success will be measured by the number of Blockfrost API calls, indicating direct on-chain data requests. Our ultimate goal would be to measure transactions submitted by our MCP server and AI agent, showcasing its adoption by users.

[Your Project and Solution] Solution

Please describe your proposed solution and how it addresses the problem

MCP Server: The Gateway to Cardano On-chain Data

Our proposed solution is to build a Model Context Protocol (MCP) server that acts as a dedicated AI-access layer for the Cardano ecosystem. At its core, the MCP server will aggregate multiple data sources: Blockfrost.io and Cardanoscan for real-time on-chain data such as wallet activity, transaction history, and governance actions; Firecrawl for targeted search and web data crawling to extend AI context; CoinMarketCap for up-to-date market pricing and trends; and Lidonation’s API for granular insights into funded projects, their milestones, and governance participation.

These endpoints will be normalized into AI-readable formats and made openly accessible, allowing not only our AI Agent but any third-party AI agent to easily connect and retrieve information. We empower the rapidly growing number of users who rely on AI agents daily, enabling them to access Cardano's on-chain and ecosystem data seamlessly.

AI Agent: Utilizing the Power of the MCP

On top of this data infrastructure, we will deliver practical user interfaces: a web portal for direct public access, an X/Twitter bot for instant social media queries, and an open-sourced codebase for integration with automation platforms such as n8n to allow no-code automation of Cardano data workflows. This architecture solves the current accessibility gap, ensuring AI agents can not only access but also reason about Cardano’s activity. In doing so, it positions Cardano as an AI-integrated blockchain ready for the automation-first internet.

[Your Project and Solution] Impact

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

Our project will transform Cardano into an AI-accessible blockchain. By making on-chain and ecosystem data instantly available to AI agents through the MCP server, we remove the barrier of information accessibility in blockchain. For the first time, AI developers will be able to build intelligent assistants, monitoring systems, and autonomous tools that natively understand Cardano. Blockchain analysts will gain the ability to run complex queries across wallets, governance actions, market movements, and Catalyst project outcomes in seconds.

As large language models evolve toward Artificial General Intelligence (AGI), having Cardano’s data seamlessly integrated into their context will make the Cardano blockchain not just accessible but an active pioneer in AI-driven economies.

[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?

Our capability is built on relevant technical expertise, a history of successfully delivering funded projects in the decentralized AI space, and proactive validation of our proposed architecture.

Our Project Team

Khresna Pandu Izzaturrahman (AI Engineer):

Linkedin: https://www.linkedin.com/in/khresna-pandu/

Github: https://github.com/KhresnaPanduI

Pandu's experience lies at the intersection of AI agents, large language models, and high-performance infrastructure, making him perfectly suited to lead the technical development of the MCP server and AI agent. He has also been an active contributor to the Cardano ecosystem, serving as a community reviewer since Fund 8 (2022).

• Direct AI Agent Experience: As a Lead AI Engineer, he has designed and deployed multiple AI agents within a microservices architecture for an enterprise-level chatbot used by over 2 million users. This work involved building complex logic that reduced customer service response times from 15 minutes to under 5 seconds (>99% reduction) and gives him direct, hands-on experience with frameworks like Langchain, n8n and AutoGen.
• Advanced Infrastructure & MLOps: Pandu has managed on-premise H100 GPU servers for high-performance LLM inference. His expertise includes fine-tuning a wide array of models (LLaMA, Mistral, Qwen) and implementing robust MLOps pipelines using Docker, Nginx for load balancing, and Prometheus/Grafana for monitoring. This ensures our project's infrastructure will be scalable, stable, and professionally managed.
• High-Performance Data Systems: He has built and scaled vector databases like Milvus to manage large knowledge bases with sub-30ms semantic search responses and has developed high-throughput FastAPI APIs capable of handling 500+ requests per minute, proving he can build the fast and responsive data backend this project requires.

Fitra Rahmani Khasyah (Full Stack Developer):

Linkedin: https://www.linkedin.com/in/fitra-rahmani

Github: https://www.github.com/khasyah-fr

Fitra is a software engineer with extensive experience in building scalable full-stack applications within e-commerce, finance, and AI industry. He managed funded projects from initial concept to complete implementation on SingularityNET, a decentralized platform powered by Cardano’s blockchain. He has been participating actively in the Project Catalyst, having served as a reviewer since Fund 8.

• Interface & API Development: Fitra has a proven track record of architecting and developing front-end interfaces and back-end services using React, Node.js and Golang. He has managed large-scale PostgreSQL and Redis clusters and built systems that orchestrate over millions of transactions securely and reliably. He will lead the development of our web interface and social media bot for users to interact with our AI Agent. He will also handle the development of endpoint calls needed for the MCP server.
• Project Management & Software Delivery: Fitra previously secured funding and delivered blockchain projects. He had hands-on experience managing timelines, deliverables, and software requirements related to the projects. Thus, he will lead the project in terms of project management and testing, ensuring high-quality standards for our solution.

Dani Ihza Farrosi (Blockchain Developer, Community & Growth):

Linkedin: https://www.linkedin.com/in/daniihza/

Github: https://github.com/danivideda

Dani is an experienced software engineer having delivered backend services using Javascript and PHP. His deep involvement in Project Catalyst since Fund 6 (2021) as a reviewer and a moderator gives him contextualized understanding of our project’s success factors.

• Developer Advocacy & Community Growth: Dani had been involved in Plutus Pioneer, Eastern Town Hall, Catalyst School, and Cardano ecosystem very early on. He will lead as our project’s developer advocate, responsible for community engagement through our social media channels (Twitter/X, Discord).
• Interface & Blockchain Development: Dani will also develop the front-end interfaces and back-end services with emphasis on blockchain fundamentals. He will assist the project as we navigate through Cardano’s on-chain data.

Proven Track Record

Our team's ability is demonstrated by our track record within other decentralized funding ecosystems:

• Sign Language Translator AI (SLTA): Our team secured a $50,000 grant from SingularityNET's Deep Funding, a funding program for novel AI services. SLTA ranked 3rd out of 91 competitive technical projects and passed rigorous review evaluation for technical feasibility and impact. For more info: https://deepfunding.ai/proposal/sign-language-translator-ai-slta/
• SingularityNet & Deep Funding: Indonesia Chapter: Our team also secured funding for and delivered a community-building initiative, proving our management skills in a project's full lifecycle, from the first milestone all the way to final reporting and handover. For more info: https://deepfunding.ai/proposal/singularitynet-and-deep-funding-indonesia-chapter/

[Milestones] Project Milestones

Milestone Title

Project Setup and Foundational Infrastructure

Milestone Outputs

- Repository & Account Initialization: Create a dedicated GitHub organization and repositories for the project. Initialize all necessary developer accounts, including Blockfrost.io, CardanoScan, CoinMarketCap, Claude, and Firecrawl, to ensure API access.

- Community Channel Setup: Launch the official Twitter/X account and Discord server to begin community engagement and provide a platform for future updates.

- Initial UI/UX Design: Develop low-fidelity wireframes and user flow diagrams for the web interface, outlining the core user journey for interacting with the AI Agent.

Acceptance Criteria

• All required external service accounts are created and verified.
• The UI/UX designs are finalized and validated for the development phase containing chat and authentication page.
• Community channels are established and have introductory content posted.

Evidence of Completion

• A public link to the GitHub repository.

• Links to the active Twitter/X and Discord channels.

• Screenshots and design files for the UI/UX wireframes.

Delivery Month

1

Cost

25000

Progress

20 %

Milestone Title

Core MCP Server and Initial Integrations

Milestone Outputs

- MCP Server Scaffolding: Develop the foundational architecture for the Model Context Protocol (MCP) server, including the API framework and core logic for handling data requests.

- Blockfrost Integration: Integrate Blockfrost.io and CardanoScan endpoints into the MCP server to enable access to real-time on-chain data, such as transaction histories, wallet balances, and staking information.

- CoinMarketCap Integration: Connect the MCP server to CoinMarketCap APIs to fetch real-time price data for ADA and other Cardano-native tokens.

- Initial AI Agent Logic: Develop the initial version of the AI agent, capable of processing queries and interacting with the integrated MCP server tools (Blockfrost and CoinMarketCap).

- Initial Front-End Development: Begin building the web interface using a modern JavaScript framework, implementing the UI layouts and components based on the finalized UI/UX design.

Acceptance Criteria

• The MCP server can successfully retrieve and structure data from both Blockfrost and CoinMarketCap APIs.
• The AI Agent can correctly answer test queries like "What is the current price of ADA?" and "Show me the latest transactions for this wallet address."
• The web interface is rendered correctly for chat and authentication features and ready for backend integration.

Evidence of Completion

• A video demonstration of the MCP server successfully fetching and processing data from Blockfrost.io, CardanoScan, and CoinMarketCap in a terminal or test environment.
• Access to the GitHub repositories for the MCP server and the front-end application.
• A document containing test queries and the corresponding successful responses from the AI Agent.
• A URL link to the front-end codebase and interface.

Delivery Month

3

Cost

30000

Progress

50 %

Milestone Title

Integration, Testing, and Beta Release

Milestone Outputs

- Advanced Data Source Integration: Enhance the MCP server by integrating Lidonation’s Catalyst Explorer API to access project funding data and Firecrawl to enable targeted web search and crawling.

- Front-End and Back-End Integration: Connect the web application front-end to the MCP server, creating the initial end-to-end user experience.

- Beta X/Twitter Bot Deployment: Develop and deploy an initial version of the X/Twitter bot. This version will be for public testing and will be able to process tagged queries and post replies using data from the MCP server.

- Deploy for Public Testing: Deploy the integrated web application and Twitter bot to a public-facing staging environment to gather crucial community feedback and identify bugs.

Acceptance Criteria

• The beta web application is live and functional for testing purposes.
• The beta X/Twitter bot is operational and can be tested publicly on the X/Twitter platform.
• The AI Agent can successfully answer simple queries that require the newly integrated tools. For example:

1. Using Lidonation: "Tell me about the list of proposals from Catalyst Fund 12."
2. Using Firecrawl: "What is the latest blog post on the IOG website?"

Evidence of Completion

• A public URL to the beta web application for community testing.
• A link to the active beta X/Twitter bot's profile, along with screenshots of example interactions.
• Updated GitHub repositories for the MCP server and web application.
• A video demonstration showcasing the AI Agent's new capabilities using Lidonation and Firecrawl on both the web platform and Twitter.

Delivery Month

5

Cost

30000

Progress

80 %

Milestone Title

Final Launch, Documentation, and Project Closeout

Milestone Outputs

- Production Deployment: Deploy the final, production-ready version of the web application and MCP server.

- Technical Documentation: Create technical documentation for the MCP server, detailing how other developers can connect their own AI agents or applications to our infrastructure.

- Community Engagement: Actively engage with the community through social channels, gather feedback, and address any bugs or issues identified during the initial launch period.

- Project Closeout Reporting: Create the final Project Closeout Report (PCR) and a project summary video (PCV) to be shared with the Cardano community, detailing the project's journey, achievements, and impact.

Acceptance Criteria

• The AI Agent is live and accessible to the public on both the website and Twitter.

• The technical documentation is clear, comprehensive, and enables a third-party developer to successfully integrate with the MCP server.

• Community updates cover all project milestones and provide clear insights into the project's delivery.

• At least 50 unique users have submitted feedback via the official form.

Evidence of Completion

• The final Project Closeout Report and Video (PCR and PCV).
• Links to the final, deployed web application and X/Twitter bot.
• A link to the public technical documentation.
• Screenshots and links to posts showcasing community updates and project announcements.

Delivery Month

6

Cost

15000

Progress

100 %

[Final Pitch] Budget & Costs

Please provide a cost breakdown of the proposed work and resources

Our total requested budget is 100,000 ADA. This funding is allocated across four key milestones, primarily covering personnel costs for our team of three, who bring extensive experience as AI and Full Stack Engineers. The budget is structured to reflect the development intensity of each phase, from foundational setup and core development to the final launch and handover.

Milestone-Based Budget Breakdown (nominal in ADA)

Milestone 1: UI/UX Design [15,000 ADA], Community Channel Setup [5,000 ADA], Repository and Account Initialization [5,000 ADA]

Milestone 2: MCP Server Development [12,000], Initial AI Agent Logic Development [10,000], Initial Front-End Development [8,000]

Milestone 3: Full Stack Integration [14,000], Twitter Bot Development & Deployment [8,000], Advanced MCP Server Integration [8,000]

Milestone 4: Production Deployment & System Optimization [11,500], Technical Documentation Creation [1,000], Project Closeout Reporting [2,500]

The detailed budget breakdown per tasks in each milestone can be seen in below google sheet:

https://docs.google.com/spreadsheets/d/1zQ8S6d0FaWPvr-PRIOHvc0x6KQXtjp-IABD83xyxgqs/edit?usp=sharing

[Final Pitch] Value for Money

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

We aim to build the first open-source Model Context Protocol (MCP) server for Cardano, making its data accessible to other AI agents like Claude, and automation platforms like n8n, so developers, AI companies, and the community can integrate tools with ease. By creating reusable infrastructure, we position Cardano as the key blockchain player in the AI era. All outputs will be open-source under Apache 2.0, inviting community contributions. Our Catalyst-experienced team has been active since Fund 6, understands community standards, and delivers at rates lower than typical market rates for experienced full-stack developers and AI engineers (50–100 USD/hour), ensuring cost-effectiveness.

References:

Freelance Full-Stack Developer: https://www.upwork.com/hire/full-stack-developers/

Freelance AI Engineer: https://www.upwork.com/hire/ai-developers/

Freelance Blockchain Developer: https://www.upwork.com/hire/blockchain-developers/

[Required Acknowledgements] Consent & Confirmation

Terms and Conditions:

Yes