- OutSystems' Agent Workbench helps enterprises and solo developers and manage agentic AI systems
- Early adopters report efficiency gains from new AI agent workflows
- General release adds marketplace, Model Context Protocol, and broader LLM compatibility
OutSystems has announced the general availability of Agent Workbench, a new platform designed to help enterprises develop and manage agentic AI systems.
The low-code development firm revealed the news at its annual ONE Conference in Lisbon, an event that brings together IT leaders, developers, and partners to explore advances in software and AI.
Agent Workbench, which was first unveiled in July, is intended to enable organizations to create and coordinate intelligent agents across different workflows and data sources.
Bringing legacy systems into the future
OutSystems says that while most enterprises are experimenting with agent-based AI, many are facing challenges related to governance, security, and integration.
“When applied strategically, agentic AI is one of the most powerful levers for enterprise innovation and transformation, empowering organizations to drive value faster, optimize workflows, and bring legacy systems into the future,” said Woodson Martin, CEO of OutSystems.
“Agent Workbench provides enterprises with the infrastructure to achieve unprecedented results with agentic systems, while maintaining critical observability and control. Our Early Access Program participants have already realized impressive gains with Agent Workbench, positioning them as industry leaders in navigating the agentic future,” Martin said.
Of the companies that have already tested the platform, Thermo Fisher Scientific used it to build a customer escalation agent that cuts manual work and speeds resolution, while The Arch Company replaced spreadsheet-driven processes with an agent that classifies and routes inquiries automatically.
In India, Grihum Housing Finance deployed agents to streamline loan underwriting and property evaluations, addressing inconsistent manual workflows.
The general release includes new features such as an agent marketplace and support for the Model Context Protocol, which connects agents with enterprise systems and external tools.
It also expands compatibility to include a wide range of large language models from major providers including OpenAI, Anthropic, Google, IBM, and others.
OutSystems, founded in Lisbon, focuses on helping enterprises build and manage applications quickly, and more recently, apply AI within software development.
Agent Workbench is available in ODC and OutSystems Personal Edition (only in testing) and costs 20K Euros; however, if you use Agent Workbench to launch an agent before the end of 2025, it will be free for the duration of an existing contact.
You might also like
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.