How Palette Built Fine-Grained Authorization for AI Agents

Palette is building a context layer for AI, and its desktop app lets non-engineers work with the agent of their choice using company context pulled in from Slack, Jira and Linear. That raises a hard question: who is allowed to see what. Some customers run private teams inside their own company that not even an org admin should see into, and information can never leak between organizations. Role-based access control is where most teams start, and it wasn't flexible enough. Tiffany Lozenski, founding product engineer at Palette, shares how the team moved to fine-grained authorization with WorkOS FGA, how permissions get checked before an agent acts, and how that collaboration turned into an FGA offering inside Mastra itself. Read the full case study: https://mastra.ai/customers/palette Watch Tiffany's talk at TypeScript AI Conference, What I Learned Shipping FGA to Production: https://www.youtube.com/watch?v=aT2viVoHs7A Palette: https://palette.team 📚 MASTRA RESOURCES Website: https://mastra.ai X: https://x.com/mastra_ai Discord: https://mastra.ai/community/discord GitHub: https://github.com/mastra-ai Learn Mastra: https://mastra.ai/learn Principles of Building AI Agents: https://mastra.ai/books/principles-of-building-ai-agents Patterns of Building AI Agents: https://mastra.ai/books/patterns-of-building-ai-agents WHAT IS MASTRA Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development, from prototype to production.