Salesforce adds Agent Evals to Agent Fabric — automated checks curb runaway AI agent costs
Context
Salesforce has expanded Agent Fabric with Agent Evals, a system that continuously evaluates AI agent performance and costs and aims to catch failures before they reach customers. The tool works across agents built on any platform, not just Salesforce's own. For marketing leaders scaling agents into production, the pitch is cost control and reliability, so it is worth checking the source for how the evaluation criteria and cost controls actually work.
Original Source
Salesforce Newsroom
Salesforce Expands Agent Fabric to Automatically Stop Runaway Costs and AI Agent Failures Before They Impact Customers
Radar items are brief signals worth monitoring