Skip to main content
Challenge

EU DMA/DSA Policy Compliance Graph Agent

Create a graph-based multi-agent system using LangGraph and Claude Sonnet 4 to assist enterprises in navigating the complexities of EU Digital Markets Act (DMA) and Digital Services Act (DSA) compliance, given the intensified enforcement. This challenge requires building agents that can interpret intricate legal texts, assess company practices against regulations, and identify potential compliance gaps. The system will employ an A2A protocol for agents to communicate and collaborate on different aspects of regulatory analysis. MCP tool integration will provide agents with access to internal company policies, platform usage data, and legal databases. Optuna will be used to optimize prompt engineering and agent workflow parameters, ensuring the most accurate and efficient interpretation of legal documents. The final output will be a detailed compliance report and actionable recommendations for mitigating risks.

Workflow AutomationHosted by Vera
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Implement a LangGraph-based multi-agent system with Claude Sonnet 4 and Optuna to automate EU DMA/DSA compliance analysis using A2A protocol and MCP tool integration.

Delivery guide

How work is evaluated

Evaluation

The evaluation will assess the system's ability to accurately identify compliance obligations and gaps based on simulated company data and extracts from the DMA/DSA. The generated compliance report will be judged on its accuracy, completeness, and the relevance of its recommendations.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Learning goals

What you should walk away with

  • Implement A2A protocol for seamless agent communication.

  • Design graph-based agent workflows using LangGraph for regulatory compliance.

  • Optimize prompt engineering with Optuna for improved LLM performance.

  • Build MCP-enabled tool integrations for legal and internal data access.

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 1 tasks

Challenge setup

The evaluation will assess the system's ability to accurately identify compliance obligations and gaps based on simulated company data and extracts from the DMA/DSA. The generated compliance report will be judged on i...

Challenge input

{'company_profile': {'name': 'GlobalPlatform Inc.', 'services': ['social media', 'online marketplace'], 'user_count': '300M EU users'}, 'dma_articl...

Agent execution

No agent tool is listed for this challenge.

Evaluated output

{'company_summary': '...', 'dma_assessment': [{'article': '...', 'status': 'compliant|non-compliant|at-risk', 'explanation': '...', 'recommendation...

Evaluation checks
  • Checks if all required sections of the compliance report for specified DMA/DSA articles are present.
  • Verifies that all specified DMA and DSA articles in the input are addressed in the output.
Configured evaluation evidence
  • ComplianceStatusAccuracy target: 85
Configured runtime
  • Python execution harness

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • Checks if all required sections of the compliance report for specified DMA/DSA articles are present.
  • Verifies that all specified DMA and DSA articles in the input are addressed in the output.

Metrics

  • ComplianceStatusAccuracy · Target 85
  • RecommendationRelevance · Target 80

Workflow tasks

  1. Task 1

    Input: {'company_profile': {'name': 'GlobalPlatform Inc.', 'services': ['social media', 'online marketplace'], 'user_count': '300M EU users'}, 'dma_articles': ['Article 5', 'Article 6'], 'dsa_articles': ['Article 25', 'Article 34'], 'internal_data_links': ['link_to_privacy_policy.pdf', 'link_to_moderation_guidelines.pdf']}

    Output: {'company_summary': '...', 'dma_assessment': [{'article': '...', 'status': 'compliant|non-compliant|at-risk', 'explanation': '...', 'recommendations': '...'}], 'dsa_assessment': [{'article': '...', 'status': '...', 'explanation': '...', 'recommendations': '...'}], 'overall_risk_score': 'high|medium|low'}

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about EU DMA/DSA Policy Compliance Graph Agent