Documentation

Connect your first agent and run a discovery evaluation in under 5 minutes.

• Connect via API endpoint
• Upload skill files
• Understanding your readiness score

Upload your skill files (.md), CLAUDE.md configuration, and supporting files. We simulate the full Claude Code environment during evaluation.

• Preparing skill files
• Mock infrastructure setup
• Using training assets

Connect any agent with an HTTP endpoint. We support OpenAI, Anthropic, and custom request formats with automatic response path detection.

• Endpoint configuration
• Authentication setup
• Test mode headers

Browse 667+ scenarios across 15 industries. Each scenario includes success criteria, failure indicators, and real-world grounding.

• Browse by industry
• Understanding difficulty levels
• Custom scenario generation