As organizations embrace agentic workflows, agents are increasingly connecting to third-party apps like Salesforce, Jira, Slack, and Notion and taking real actions. That shift creates a new security and governance gap: traditional identity and access models were designed for humans and services, not autonomous tool-calling agents. In this lightning talk, we’ll break down the core building blocks of agent access governance: agent identity, scoped permissions per tool/action, step-up and approval for high-risk operations, and complete auditability of every tool call. You’ll leave with a practical framework you can apply immediately to ship agentic workflows safely across your SaaS ecosystem.