Agents Are Designed to Be Autonomous Problem-Solvers,Possibly Communicating With Other Agents and Users, and Are Therefore Equippedwith Sufficient Cognitive Abilities to Reason About a Domain, Make Certaintypes of Decisions Themselves, and Perform the Associated Actions. In Thispaper, We Propose to Integrate Agents In a Cooperative Intelligent Decisionsupport System. the Resulting System, Called Macids Is Designed to Supportoperators During Contingencies. During the Contingency, the Operators Usingmacids Should Be Able To: Gather Information About the Incident Location;Access Databases Related to the Incident; Activate Predictive Modelingprograms; Support Analyses of the Operator, and Monitor the Progress of Thesituation and Action Execution. In Macids the Communication Support Enhancescommunication and Coordination Capabilities of Participants. a Simple Scenariois Given, to Illustrate the Feasibility of the Proposal. We Are Investigating Techniques For Developingdistributed and Adaptive Collections of Information Agents That Coordinate Toretrieve, Filter and Fuse Information Relevant to the User, Task and Situation,As Well As Anticipate User's Information Needs. In Our System of Agents,Information Gathering Is Seamlessly Integrated With Decision Support. the Taskfor Which Particular Information Is Requested of the Agents Does Not Remain Inthe User's Head But It Is Explicitly Represented and Supported Through Agentcollaboration. In This Paper We Present the Distributed System ...