Salesforce is rolling out new features in Agentforce 3 aimed at helping solution providers better manage and scale AI agent deployments, reports CRN. With native support for Anthropic’s Model Context Protocol (MCP) launching in July and a new Command Center set to follow in August, the platform is evolving to give partners more control, visibility, and interoperability.These additions reflect Salesforce’s broader effort to help its ecosystem support a growing number of AI-driven use cases across regulated industries and multi-cloud environments. The Command Center stands out as a practical tool for managing the lifecycle and performance of AI agents. Solution providers will be able to monitor metrics like agent latency, escalation frequency, and error rates, while integrating observability tools such as Datadog and Splunk.Built directly into Agentforce Studio, the feature is designed to help partners deliver more value-added services such as tuning KPIs and supporting real-time troubleshooting.Also in the pipeline are new enhancements around MCP and A2A integration, including connectors from MuleSoft to make APIs agent-ready, with governance features like activity tracing and traffic controls. Heroku AppLink is being introduced to streamline custom MCP server deployment, while Anthropic Claude models hosted inside the Salesforce trust boundary offer more secure AI model access for customers in sensitive sectors.
Related Events
You can skip this ad in 5 seconds