[Spidex] AI-Powered Risk Analysis Tool for Cardano DeFi

[Proposal setup] Proposal title

Please provide your proposal title

[Spidex] AI-Powered Risk Analysis Tool for Cardano DeFi

[Proposal Summary] Budget Information

Enter the amount of funding you are requesting in ADA

89615

[Proposal Summary] Time

Please specify how many months you expect your project to last

7

[Proposal Summary] Translation Information

Please indicate if your proposal has been auto-translated

Yes

Original Language

vi

[Proposal Summary] Problem Statement

What is the problem you want to solve?

Lack of a comprehensive, easy-to-understand risk assessment tool makes Cardano DeFi users vulnerable to significant asset losses, and creates a gap to fill to protect and strengthen ecosystem trust.

[Proposal Summary] Supporting Documentation

Supporting links

• https://app.spidex.ag/
,
• https://spidex.ag
,
• https://spidex-ai.gitbook.io/spidex-ai-docs
,
• https://x.com/Spidex_ag
,
• https://discord.com/invite/xxxrzcwNfB

[Proposal Summary] Project Dependencies

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

Yes

Describe any dependencies or write 'No dependencies'

Spidex Sentinel depends on multiple external sources, including Cardano node APIs, DEX APIs, data streams, and third-party AI services.

[Proposal Summary] Project Open Source

Will your project's outputs be fully open source?

No

License and Additional Information

Spidex AI will release selected open-source components for community review and contribution, including: anonymized on-chain datasets, public technical documentation, and selected analysis tools.

The trained AI models, internal databases, and certain critical backend components will remain proprietary and closed to ensure security and prevent unauthorized exploitation.

In the future, we will consider publishing some additional components.

[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).

Spidex Sentinel is the first AI-powered, multi-layer risk analysis dashboard dedicated to the Cardano DeFi ecosystem. Unlike existing tools that focus on isolated data sets or raw blockchain metrics, Sentinel combines:

• AI-based Smart Contract Auditing for Plutus code
• Market & Liquidity Risk
• Community & Sentiment Intelligence

These layers are integrated into a single composite risk score and embedded directly into Spidex AI’s chatbot to receive actionable insights instantly — a leap forward in accessibility, depth, and real-time adaptability for Cardano DeFi participants.

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

The MVP will showcase a fully functional Spidex Sentinel integrated into the Spidex AI app includes:

• AI trained on Plutus code to detect vulnerabilities and check mint/honeypot risks.
• Assesses liquidity, holder,... (this feature already exists in Spidex AI and will be enhanced for greater accuracy and effectiveness).
• NLP scans X and Telegram to detect narratives, track Attention Index, evaluate KOLs, and identify bots/FUD.

Currently, Spidex AI core chatbot and data processing pipeline are operational. The Sentinel risk analysis tool planned for the first step.

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

• Users: Reach 2,000+ active users engaging with the Sentinel feature within 3 months of launch.
• Queries: Process 5,000+ risk analysis requests via chatbot in the first quarter.
• Coverage: Automatically scan at least 80% of newly created tokens on Cardano within 24 hours of deployment.
• Accuracy: Achieve >85% accuracy in identifying high-risk tokens compared to independent audit reports.
• Ecosystem impact: Significantly reduce the number of rug pulls or exploits among tokens flagged by Sentinel.

[Your Project and Solution] Solution

Please describe your proposed solution and how it addresses the problem

A. PROBLEM

We have identified three key issues that are directly impacting the user experience and safety in Cardano DeFi:

1. Liquidity and capital on the Cardano network are declining faster than the market average

According to Messari’s Q1/2025 report, Cardano’s TVL dropped by 29% compared to the previous quarter, down to $319.3 million, while the average decline across the DeFi market was 26%.

(Source: Messari) Cardano TVL Chart – Q1/2025

Impact: The TVL decline exceeding the market average indicates that Cardano is facing stronger capital outflows, leading to higher liquidity risks, greater price volatility, increased susceptibility to manipulation even with small trading volumes, and a heightened risk of large-scale sell-offs.

(Source: Messari) Cardano Metrics Overview

2. Investors are shifting towards high-risk trading without adequate analytical tools

Daily trading volume on Cardano DEXs rose from 7% in 2024 to 39% of the total market in June 2025. Notably, a large portion of this activity is concentrated in highly volatile memecoins.

(Source: Congape) DEX Volume Growth Chart

Impact: Investors face greater risks from highly volatile tokens, yet current tools do not provide sufficient actionable insights, early warnings, or in-depth analysis to help protect capital.

3. Difficulty in tracking community, media, and social behavior signals quickly and effectively

DeFi-related information on platforms like X, Telegram, and community forums is often chaotic, containing shilling, FUD, and misinformation, making it easy for investors to be manipulated.

(Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5144847))

Impact: urrent analytics in the Cardano ecosystem are unable to automatically detect emerging trends, measure attention, assess KOL credibility, or cover a broad range of datasets/assets. This creates information asymmetry, putting retail investors at a disadvantage compared to early trend adopters.

B. SOLUTION: Spidex Sentinel - AI-Powered Risk Analysis Tool for Cardano DeFi

To address these challenges, we propose Spidex Sentinel – a comprehensive risk analytics dashboard integrated directly into the Spidex AI application, built on three core analytical pillars:

1. Smart Contract Analysis (AI-powered Audit)

We will develop AI auditing capabilities trained specifically on Plutus code (leveraging OpenAI, Llama, Claude), enabling the system to analyze complex contract structures and automatically detect patterns linked to common vulnerabilities.

• Plutus Contract Scan: Automatically detect known security vulnerabilities and exploitable functions.
• Mint Policy & Honeypot Analysis: Verify whether the token mint policy is locked to prevent uncontrolled inflation and detect potential honeypots.

2. Market & Liquidity Risk Analysis

This module evaluates the financial stability of a token. It extends Spidex AI’s existing capabilities, enhancing precision and early-warning functions for investors.

• Liquidity Analysis: Check liquidity lock status and duration.
• Holder Concentration Analysis: Flag tokens where a small number of wallets hold a large share of supply, increasing the risk of price manipulation.
• Dump Risk Radar: Assess unrealized profits of traders to predict selling pressure.

3. Community & Social Risk Analysis

This is our key differentiator. We will deploy advanced Natural Language Processing (NLP) models to scan platforms like X, Telegram, and Cardano community forums via key APIs to extract relevant keywords and analyze thousands of discussions to:

• Story Detection: Identify emerging narratives or unsustainable hype campaigns (shilling).
• Attention Index: Track mention frequency and sentiment (positive/negative) across social channels.
• KOL Evaluation: Automatically identify KOLs discussing the token and assess their credibility based on historical performance, while also detecting sudden spikes in bot activity or FUD campaigns.

C. Demo & How It Works

Step 1 — Ask

The user searches for a token on the Spidex Chatbot using natural language: “How does Spidex Sentinel evaluate token X?”

Step 2 — Analyze (3 Layers)

• Layer 1 · AI Smart Contract Audit (Plutus)
• Layer 2 · Market & Liquidity Risk
• Layer 3 · Community & Social Risk (NLP)

Step 3 — Output

• The AI responds in natural language with a consolidated risk score (1–100, color-coded & Positive / Neutral / Negative label), and a detailed report including explanations and visual evidence.

Through this approach, Spidex Sentinel directly addresses the problems of information asymmetry and risk in DeFi, delivering Cardano users a fully evaluated, automated, easy-to-understand, and seamless information experience within Spidex AI.

C. IMPLEMENTATION ROADMAP

1. Master Plan

To ensure effective execution, we have defined four main phases for the project:

(More Details: Milestones Overview)

Within the scope of this proposal, we will focus from Milestone 1 (Details please view in the Implementation Plan).

We will maintain a detailed 7-month Master Plan, with all milestones aligned to this timeline.

(More Details: Master Plan)

This Master Timeline will serve as the primary reference for tracking progress, ensuring that every workstream from product development and AI model training to marketing campaigns and evaluatio, remains on track and fully synchronized.

2. Statement of Milestones (SOM) Preparation

In alignment with Catalyst Fund 14 guidelines, our Statement of Milestones (SOM) is prepared in advance, with input and support from Cardano Ambassador Mrs Ha. This approach ensures:

• Transparency & Clarity – The plan is fully aligned with Catalyst standards, making our project roadmap clear and traceable for all stakeholders.
• Ease of Review – Community Reviewers (CR) can quickly assess the deliverables, timelines, and budget allocation for each milestone.
• Performance Assurance – Our projected completion percentage for each milestone consistently exceeds the corresponding budget percentage, demonstrating delivery efficiency and accountability.

(More Details: SOM)

Furthermore, our final milestone meets the Catalyst minimum threshold of 15% of total project scope, achieving 20%, which provides additional assurance of sustained delivery until project close-out.

[Your Project and Solution] Impact

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

A. Quick Comparison

Currently, monitoring tools such as Xerberus support Cardano but remain focused on basic analytics and have not yet integrated AI for automated risk assessment. Meanwhile, tools on BSC such as Dexcheck.AI provide advanced risk analysis but do not serve the Cardano community, creating a gap that Spidex Sentinel is designed to fill.

Table: Quick Comparison – Current Tools vs. Spidex Sentinel (SWOT)

Compared to existing tools, Spidex Sentinel is the pioneering solution for Cardano, integrating AI to detect risks quickly, providing reliable data, and enabling users to make smarter investment decisions.

B. IMPACT

Spidex Sentinel will generate measurable and positive impact across the Cardano ecosystem, aligning with the core evaluation criteria of Catalyst.

• For users: Minimize capital loss from scams/honeypots and enhance the ability to self-assess before trading.
• For the Cardano ecosystem: Improve DeFi trust and safety, attract new users, reduce scam incidents that damage Cardano’s reputation, and potentially increase Total Value Locked (TVL).
• For developers: By open-sourcing the core models, we provide a valuable resource for other DApp builders on Cardano, enabling them to integrate or build on our technology to strengthen the security of their own applications.

C. Projected Quantitative Impact Metrics

1. Coverage & Technical Performance

• Sentinel is expected to automatically scan at least 80% of new Cardano tokens within 24 hours of creation, reaching >95% coverage by the second year.
• Sentinel’s AI model is projected to achieve >85% accuracy in classifying risky vs. safe tokens after the training phase.

2. Ecosystem Risk Reduction

• Reduce approximately 20% of scam tokens in Cardano DeFi during the first year through fast scanning and automated risk analysis.

3. Adoption & User Growth

• Increase 15% in active DeFi users within six months due to enhanced trust and transparent decision-support tools.
• Generate >500 chatbot risk analysis queries per month.
• Achieve 80% positive feedback rating from users on tool usefulness and reliability.

4. Ecosystem Impact (Cardano-wide)

• Contribute to a 3–5% increase in Cardano TVL in the first year by reinforcing user confidence.

Result sharing: Publish monthly aggregated data and public risk reports for the Cardano community.

D. Sentinel Demo

As Sentinel’s features are not yet live, we will showcase both a demo version of Spidex Sentinel and reference data from Dexcheck.AI on BSC to illustrate how AI can detect risky tokens, issue alerts, and support safer decision-making. This demonstrates the real-world impact Sentinel can bring to the Cardano ecosystem.

Demo: Spidex Sentinel (current mockup)

Illustration: Dexcheck.AI on BSC

[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?

The Spidex AI team possesses the full range of skills and expertise required to execute the project effectively.

1. Strengths & Professional Expertise

• AI & Machine Learning: Proficient in building AI models based on existing AI frameworks such as OpenAI, Lama, GPT, and Grok X to analyze smart contract code structures and develop anomaly detection models for on-chain data.
• Blockchain Development: Experienced in collecting and processing real-time data from the Cardano blockchain through features already integrated into Spidex AI, such as liquidity analysis, top holders, token distribution, and transaction behavior.
• DeFi Product Development: Successfully deployed the Spidex AI On-chain Analysis Chatbot, demonstrating the capability to combine AI technology with products that directly serve Cardano users.

2. Team Members

• AI Engineers: 5+ years of experience in training GNN, anomaly detection, and NLP models.
• Blockchain Developers: 4+ years of experience developing dApps and integrating Cardano data.
• Project Managers: 8+ years of experience coordinating technology and AI product development projects.
• Marketing Managers: 4+ years of experience in blockchain marketing, collaborating with industry KOLs, and managing blockchain project communities.
• Product Managers: 6+ years of experience in managing tech products, creating roadmaps, and coordinating technical teams.
• Operations Managers: 5+ years of experience in running technology projects, optimizing processes, and managing resources.
• Graphic Designers: 4+ years in creating branding, UI/UX, and marketing visuals for blockchain, AI and technology products.

3. Transparent Management Process

• Clear Accountability: The Project Manager oversees overall coordination, the Lead Engineer supervises code and AI model quality, and Data Specialists ensure data accuracy.
• Monitoring Process: Each milestone includes progress reports, test results, and proof of completion before moving to the next phase.
• Public Reporting: Monthly updates will be shared with the Catalyst community, including progress reports, test results, and transparent budget usage by category.

4. Risk & Mitigation Plan

To ensure the successful development and deployment of Spidex Sentinel on the Cardano network, we have identified key technical and operational risks along with clear mitigation and backup strategies. The following Risk & Mitigation Plan outlines how we will proactively manage these challenges.

(More Details: Risk & Mitigration Plan)

We are committed to delivering Spidex Sentinel on time, with high quality, and with real value for the Cardano community. This confidence comes from strong technical expertise, an experienced team, and a transparent development process. In addition, the system is designed with contingency measures and redundancy across data sources, nodes, and AI models, ensuring resilience and reliability under real-world conditions. As a result, Spidex Sentinel minimizes downtime, preserves data integrity, and strengthens long-term trust within the ecosystem.

[Milestones] Project Milestones

Milestone Title

Data Collection & AI Model Training

Milestone Outputs

• Core Dataset

     - Labeled dataset of Plutus smart contracts categorized as Positive / Neutral / Negative based on security and risk levels.
  
     - Dataset of Cardano tokens including on-chain metrics (liquidity, volume, holders, age, TVL, market cap).
  
     - Sentiment dataset from Twitter/Telegram posts about tokens, with trustworthiness labels.
  

• UI/UX design created in Figma.

• Feature Description Table detailing all platform features.

Acceptance Criteria

• Dataset includes at least 1,000 Plutus contract samples, 10,000 Cardano tokens with complete on-chain metrics, and 5,000 labeled social media posts.
• UI/UX design approved internally.
• Feature Description Table finalized and aligned with the approved UI/UX design.

Evidence of Completion

• Public link to datasets (CSV/XLSX) on Spidex AI Sheets.
• Figma design link accessible to reviewers (read-only mode).
• PDF of Feature Description Table uploaded to Spidex AI Docs

Delivery Month

3

Cost

26885

Progress

30 %

Milestone Title

Marketing Campaign & Public Beta Launch

Milestone Outputs

• Feature Release Notes describing new features, test scope, and user instructions.
• Marketing materials including social media posts, demo videos, banners, and infographics.
• Public Beta launch on app.spidex.ag.

Acceptance Criteria

• Release Notes are published on official channels (X, Discord, website).
• Marketing campaign is deployed across at least 3 channels (X, Discord, Telegram) reaching a minimum of 10,000 impressions with an engagement rate ≥ 5%.
• Public Beta maintains ≥ 99% uptime for 2 consecutive weeks, with no P1/P2 issues remaining 7 days after reporting.

Evidence of Completion

• Screenshot/PDF of Release Notes published on official channels.
• Marketing report (PDF) with list of channels, reach, and engagement metrics.
• Public Beta demo video.

Delivery Month

5

Cost

26885

Progress

60 %

Milestone Title

Platform Evaluation & Improvement

Milestone Outputs

• Public Launch of Spidex Sentinel v1.0 on app.spidex.ag, including at least 3 core features:

             (1) AI-powered Smart Contract Audit
  
             (2) Market & Liquidity Risk Analysis
  
             (3) Community & Social Risk Analysis
  

• User Feedback Report: collection of at least 200 valid user responses within the first month after launch.

• Improvement Plan v1.1: roadmap document updated based on usage data and feedback.

Acceptance Criteria

• Official public launch of Spidex Sentinel v1.0 on app.spidex.ag.
• Collection and analysis of ≥200 user feedback entries.
• Improvement Plan v1 .1 finalized and published publicly on Spidex AI Docs.

Evidence of Completion

• Public link to app.spidex.ag with v1.0 release.
• PDF report of user feedback (200+ entries).
• Improvement Plan v1.1 published on Spidex AI Docs.

Delivery Month

6

Cost

17923

Progress

80 %

Milestone Title

Project Close-out & Handover

Milestone Outputs

• Final Close-out Report (PDF) including:

    + Achievements compared to KPI targets,
  
    + 3-month post-launch Risk Analysis Report,
  
    + Impact evaluation on the Cardano community,
  
    + Proposals for next development phase (v2.0).
  

• Project Close-out Video (10–15 minutes): introduction, feature demo, results, and v2.0 roadmap.

• Public Repository: all project documentation (docs, reports, related source code) published on Spidex AI Docs.

Acceptance Criteria

• Report and video published on the official X account @Spidex_ag and Spidex AI Docs.
• 3-month post-launch risk analysis report published publicly with at least 3 real case studies.
• All previous milestones completed and accepted at 100%.

Evidence of Completion

• YouTube link to the project close-out video (10–15 minutes).
• PDF link to the Final Close-out Report (including 3-month Risk Analysis Report).
• Public repo/documentation fully accessible on Spidex AI Docs.

Delivery Month

7

Cost

17923

Progress

100 %

[Final Pitch] Budget & Costs

Please provide a cost breakdown of the proposed work and resources

Total budget proposed in this proposal: 89615 ADA

(2% risk management of over total budget)

Budget rate: ADA = 0.80, referenced as of the proposal submission date (August 11, 2025).

This rate is not based on the spot market price, but rather calculated from the last 6-month average, determined by taking (highest ADA price + lowest ADA price) / 2.

This method ensures stability and fairness in budget estimation, while also mitigating risks from short-term market volatility.

(More Details: Budget Caculation)

1.Budget allocation

(More Details:Budget Allocation)

• 50% for product development
• 30% for marketing activity
• 18% for personnel expense.
• 2% for Risk management

2. Personnel expenses

With the desire to have the opportunity to contribute to supporting Cardano, Catalyst to grow stronger, we will strive to work with a salary 14-57% lower than the average salary in the market. (please see salary benchmarking below)

[Final Pitch] Value for Money

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

The cost of the Spidex Sentinel project is considered reasonable and cost-effective when compared both to local engineering salaries and to similar AI audit projects on other blockchains:

1. Salary Benchmarking in the local country

• We keep 14%-57% savings on personnel cost for the team member's hourly rate

(More Details: Salary benchmarking)

2. Benchmarking Against AI Audit Projects on Other Chains

• Developing an AI-powered smart contract auditing tool on Ethereum typically ranges from $80,000 – $200,000 (approximately 150,000 – 300,000 ADA), depending on complexity.

(Source IdeaUsher)

(SourceSoftstack)

• Platforms such as ChainGPT are estimated to cost $80,000 – $140,000 to build with full AI features and management dashboard (IdeaUsher).
• Traditional (non-AI) smart contract audits already cost $20,000 – $100,000, and can exceed $250,000 for complex DeFi protocols.

(Source Zealynx)

(Source SecureApp)

With a total budget of 89,615 ADA, Spidex Sentinel achieves approximately 40%–70% savings compared to similar AI audit projects, thanks to:

• Leveraging the existing Spidex AI infrastructure, avoiding the need to build from scratch.
• A highly skilled engineering team in Vietnam, offering competitive costs compared to other regions.

Overview

• Our budget Allocation mainly focused on Product development (50%), Marketing Activity Expense (30%)
• Consider the team allocation at different levels of the project org chart
• Cardano Ecosystem is an emerging industry, and we need to have the proper induction and training for the new members.
• Risk management (2%)

This proposal brings exceptional value to the Cardano treasury. We are not only building a feature for a single application, but also creating a fundamental public good that enhances security for the entire DeFi ecosystem.

• High ROI for the Community: This investment has the potential to prevent millions of ADA in losses for users by helping them avoid scam projects. The return on investment, in terms of protecting community assets, is substantial.
• Driving Sustainable Growth: By making DeFi safer, we help attract and retain new users, contributing to the long-term and sustainable growth of Cardano, benefiting all ADA holders.
• Cost Efficiency: We are building on top of the existing Spidex AI infrastructure, which significantly reduces costs and development time compared to starting from scratch. The requested budget is reasonable and primarily focused on the technical expertise needed to build complex AI models.

Investing in Spidex Sentinel is not just an investment in a product, but an investment in the trust, security, and reputation of the entire Cardano ecosystem.

[Required Acknowledgements] Consent & Confirmation

Terms and Conditions:

Yes