Salesforce Expands Agent Fabric to Automatically Stop Runaway Costs and AI Agent Failures Before They Impact Customers
Salesforce adds automated agent evaluation and cost controls to Agent Fabric control plane
“What excites us most about Agent Evals is that it doesn't stop at telling us something is wrong. When an agent's quality drops, Agent Fabric can step in automatically, alerting the right owner or pulling the agent offline before a small issue becomes a widespread customer problem.”— Salesforce
Salesforce announced Agent Evals, API Evals, and advanced cost controls for Agent Fabric, its AI control plane handling 1.2 billion daily requests. The additions enable automated quality evaluation, intervention triggers including an 'Agent Kill Switch,' and hard spending caps with claimed 75% token reduction via caching and compression. This matters because it addresses a critical enterprise gap—fewer than half of organizations evaluate agents on live traffic—and positions Salesforce as a cross-platform governance layer as multi-agent deployments scale.