AI Tool Call Card

What it's for

• Agent run console where each tool call shows collapsible parameters and results.
• LLM observability panel with the duration of calls to external APIs and error inspection.
• Tool sandbox where the developer can retry failed runs or copy payloads.

For interactive editing of tool schemas is required instead of run inspection, use AI Tool Registry. For a global timeline of multiple parallel agents is required without focus on the individual call, use AI Agent Run Trace.

Includes

• Collapsible input
• Collapsible output
• Status indicator
• Duration tracking
• Copy payload
• Error recovery
• Empty state
• Keyboard navigation

Works with

• Radix UI primitives (the registry variant it ships in)
• React 19, Next.js App Router and Tailwind CSS 4, installed with the shadcn CLI
• Your theme tokens, in light and dark mode
• 1 template or blueprint on this site use it (below)