Tutorials

• Agent Annotation Benchmark: Turn annotated simulation runs into reusable benchmarks that drive Maestro config and structure optimization.
• Persona Set: Curate and deploy persona collections, then bind them to simulated entry points for richer user context.
• Mock MCP Server: Spin up platform-hosted mock MCP servers so agents can exercise tool calls without hitting production services.
• Custom Evaluator: Implement a bespoke Evaluator subclass that scores agent outputs with custom logic and metadata requirements.