Cohere Launches Security-Enhanced Agent Framework with Access Controls
Cohere has released an updated version of its North agent harness designed for enterprise use, incorporating improved security, access control, and token management features. The platform introduces skills, automations, and libraries to standardize agent behavior and ensure consistent access to shared knowledge bases and business applications. These tools aim to address growing concerns about the safety and controllability of autonomous AI agents in production environments.
Cohere's North 2 framework addresses a critical gap in enterprise AI deployment by introducing standardized procedures and governance mechanisms. The platform allows organizations to document optimal agent behaviors as reusable skills, share common knowledge through libraries, and maintain consistency across autonomous processes. By enabling granular access controls and budget tracking, the system helps prevent unauthorized data access while providing visibility into resource consumption across departments. This modular approach lets enterprises adapt the framework to existing workflows without requiring wholesale infrastructure changes.
The release reflects intensifying competition to establish trusted agent platforms for business environments, potentially affecting how organizations balance automation benefits against operational control. Companies adopting such frameworks may reduce manual processes and decision inconsistency, though they could also become dependent on proprietary governance systems. Industries handling sensitive data—finance, healthcare, legal—may find particular value in access restrictions and audit trails, while smaller enterprises might face barriers to deployment. The emphasis on sovereign deployment options suggests enterprises retain meaningful influence over where and how their AI agents operate, though actual implementation complexities remain unclear.