Configure an AI agent with purpose, tools, instructions, knowledge, triggers, output format, and test runs before publishing. This example brings Instructions and Tools together so AI product teams can find the next task without losing context.
It includes representative records such as System instructions and CRM search tool. Use the screenshot to study the layout, then adapt its components and product states to your own design system.

Builder
Publish agent
Builder
Default, loading, empty, error, permission, success
Configurable product UI theme
Product teams need to publish agent while keeping tools, knowledge sources and exceptions visible. The interface should show what changed and who should act next.
Instructions → Tools → Publish agent → Review outcome
Builder, Instructions, Tools, contextual actions, error recovery
Show constraints before adding blocks, keep a usable preview and protect unsaved changes as people test a different direction.
Pay attention to instructions and tools. People need to distinguish System instructions from CRM search tool. Keep tools visible, especially when a record changes owner or status.
Before this screen reaches handoff, verify keyboard focus, empty and permission states, and what happens after someone chooses “Publish agent”. Make error recovery as clear as the happy path.
Related screen examples to explore next: AI Chat Workspace, Tool Connector Setup, Knowledge Source Manager.
Design a agent builder screen inside our existing product. Use our current components, tokens and content language. Include Instructions and Tools views, credible records like System instructions, and a clear “Publish agent” action. Use our product permissions and show loading, empty, error and success states. Keep the output editable in Figma.