BAD MONKEY AI / PRESS RELEASE / OCTOBER 6, 2026
Bad Monkey AI Debuts AgentSpaces:
Open-Source Coordination for AI Agents
Bad Monkey AI’s first open-source product turns independently developed AI agents into coordinated teams, reducing duplicate work and making automation more reliable.
Washington, DC, October 6, 2026 — Bad Monkey AI today announced AgentSpaces, an open-source software library that enables AI agents to coordinate tasks, share results, and recover unfinished work when a worker fails. Developers can add AgentSpaces to existing applications across cloud, data center, and edge environments, reducing the cost and complexity of operating teams of agents.
As organizations deploy more AI agents, they need those agents to work together reliably and accountably. When agents built by different teams lack a common way to share results, organizations can end up paying for the same research, analysis, and data requests more than once.
AgentSpaces enables agents to discover one another and coordinate tasks through peer-to-peer communication, without requiring a central coordination server. Sharing results can reduce repeated requests to AI models and data services. When a worker fails, its unfinished task can become available for another worker to complete.
“AgentSpaces gives developers a common way to bring agents built by different teams into the same workflow,” said Albert Baker, Chief Product Officer of Bad Monkey AI. “they can add coordination to existing applications and spend more time on what their agents need to accomplish.”
“We’ve worked on peer-to-peer systems since our time at Bell Labs,” said Scott Schell, Chief Technology Officer of Bad Monkey AI. AgentSpaces applies proven ideas from distributed computing, including concepts from Sun Microsystems’ Jini and JXTA, to the coordination needs of today’s AI agents.”
Agents can continue coordinating locally when a site’s connection to the wider network is interrupted. Security features include end-to-end encryption and signed records that identify which agent produced information or took an action, helping organizations maintain traceability and accountability as they expand automation. AgentSpaces supports Java and Python, with integrations for the Spring and Embabel frameworks used by enterprise developers.
“By reusing shared results, agents can avoid duplicate requests to AI models and data services, reducing token usage and associated model costs,” said Dr. Carlton Reeves, Chief Executive Officer of Bad Monkey AI. “AgentSpaces lets organizations start with one workflow and build from there. Making it open source gives them the freedom to adapt the technology to their operational needs.”
AgentSpaces supports AI agents alongside conventional software services, robots, and sensors, sharing information and coordinating tasks through a common software layer. The open-source core is available under the Apache License 2.0 at https://github.com/badmonkeyai/AgentSpaces. For more information, visit badmonkey.ai/agentspaces. Open-source users can participate on GitHub and get help at oss@badmonkey.ai.
Bad Monkey AI will attend the 2026 AUSA Annual Meeting & Exposition at the Walter E. Washington Convention Center in Washington, D.C., October 12–14, 2026. The team will be available to demonstrate AgentSpaces and other Bad Monkey AI solutions and discuss potential applications with attendees. To request a demonstration or schedule a meeting, contact sales@badmonkey.ai.
About Bad Monkey, Inc.
Headquartered in Washington, D.C., Bad Monkey, Inc. develops enterprise software that turns real-time information into coordinated action across people, AI agents, and unmanned systems. The company’s products deliver intelligent event processing, AI agent orchestration, distributed communications, and command-and-control capabilities. For more information, visit badmonkey.ai.
Media Contact
PR@badmonkey.ai 302-402-2500
END OF RELEASE · ORIGINAL PDF AVAILABLE TO DOWNLOAD