Geopolitical Risk Agent System with Gemini 2.5 & CrewAI A2A
There is a critical need for enterprises to proactively assess and mitigate supply chain and market risks. This challenge involves building a sophisticated multi-agent system designed to monitor global events, analyze their potential impact on specific industries, and generate actionable risk reports. The system will leverage advanced LLMs for deep reasoning and agentic frameworks for collaborative analysis, helping businesses navigate complex international landscapes. Participants will develop a CrewAI-based team of specialized agents that collaborate using an A2A protocol. This system will ingest real-time news, government reports, and trade data. It will then perform extended thinking to evaluate scenarios, identify potential policy changes, and forecast economic impacts, providing a comprehensive risk assessment for a hypothetical multinational corporation.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a CrewAI multi-agent system utilizing A2A protocol, Gemini 2.5 Pro, and Claude Opus 4.1 for real-time geopolitical risk analysis and strategic reporting.
How work is evaluated
The evaluation will focus on the system's ability to accurately identify geopolitical risks, provide coherent analysis, and generate actionable recommendations. It will be judged on the correctness, completeness, and structure of the generated risk reports, as well as the effective collaboration between agents.
Challenge datasets
Public datasets, plus the private datasets you have access to.
What you should walk away with
Leverage Gemini 2.5 Pro for deep geopolitical scenario analysis.
Generate comprehensive, actionable geopolitical risk reports.
Integrate A2A communication for seamless agent-to-agent collaboration.
Implement tool integration for real-time data ingestion and RAG.
Design and implement a role-based multi-agent system using CrewAI.
Reference links and supporting material
Sample data for 1 tasks
Challenge setup
The evaluation will focus on the system's ability to accurately identify geopolitical risks, provide coherent analysis, and generate actionable recommendations. It will be judged on the correctness, completeness, and...
Challenge input
JSON object with `scenario_description` (string) and `target_industry` (string).
Agent execution
No agent tool is listed for this challenge.
Evaluated output
JSON object with `risk_summary` (string), `identified_risks` (array of strings), `potential_impacts` (array of strings), `mitigation_strategies` (a...
- Ensures the output is a well-formed JSON object.
- Checks for the presence of all required keys in the output.
- Verifies that agent-to-agent communication via the A2A protocol was detected and logged during execution.
- Analysis Completeness target: 0.9
- 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
- Ensures the output is a well-formed JSON object.
- Checks for the presence of all required keys in the output.
- Verifies that agent-to-agent communication via the A2A protocol was detected and logged during execution.
Metrics
- Analysis Completeness · Target 0.9
- Recommendation Actionability · Target 0.85
- Confidence Score Range · Target 0.8
Workflow tasks
- Task 1
Input: JSON object with `scenario_description` (string) and `target_industry` (string).
Output: JSON object with `risk_summary` (string), `identified_risks` (array of strings), `potential_impacts` (array of strings), `mitigation_strategies` (array of strings), `confidence_score` (float between 0 and 1).
Execution harness: Python
Recipe freeze marker
No recipe freeze time is recorded.