# Cascade

## Docs

- [Evaluation SDK](https://docs.runcascade.com/agent-performance/evals.md): Programmatic evaluation of agent performance with Cascade SDK
- [Python SDK Reference](https://docs.runcascade.com/agent-performance/sdk-reference.md): Complete reference for the more in depth Cascade Python SDK
- [Sessions & Tracing](https://docs.runcascade.com/agent-performance/tracing.md): Capture the full execution tree of your agent with hierarchical traces
- [TypeScript SDK](https://docs.runcascade.com/agent-performance/typescript-sdk.md): Capture agent traces with Cascade's Node.js TypeScript SDK
- [Rubrics & Evaluation](https://docs.runcascade.com/core-concepts/evals.md): Define what good looks like and automatically score every agent execution
- [Failures](https://docs.runcascade.com/core-concepts/failure-modes.md): Surface and diagnose every evaluation failure across your agents
- [Production monitoring](https://docs.runcascade.com/core-concepts/production-monitoring.md): Run evaluations at scale with Backtest and Active Evals
- [Agent Rules](https://docs.runcascade.com/developers/agent-rules.md): Integrate Cascade seamlessly with Claude Code and Cursor
- [Cascade MCP Server](https://docs.runcascade.com/developers/mcp-server.md): Manage rubrics, run evaluations, inspect traces, and debug failures directly from your IDE
- [Introduction](https://docs.runcascade.com/index.md): The autonomous evaluation and safety platform for AI agents
- [Claude Agent SDK](https://docs.runcascade.com/integrations/claude-agents.md): Auto-instrument the Claude Agent SDK with Cascade
- [LangGraph / LangChain](https://docs.runcascade.com/integrations/langgraph.md): Auto-instrument LangGraph and LangChain agents with Cascade
- [OpenAI Agents SDK](https://docs.runcascade.com/integrations/openai-agents.md): Auto-instrument OpenAI Agents with Cascade
- [Vercel AI SDK](https://docs.runcascade.com/integrations/vercel-ai-sdk.md): Automatically track Vercel AI SDK calls with full input/output logging, token usage, and tool executions
- [Anthropic](https://docs.runcascade.com/model-providers/anthropic.md): Trace Anthropic LLM calls with Cascade
- [OpenAI](https://docs.runcascade.com/model-providers/openai.md): Trace OpenAI LLM calls with Cascade
- [OpenRouter](https://docs.runcascade.com/model-providers/openrouter.md): Trace OpenRouter LLM calls with Cascade
- [Quickstart](https://docs.runcascade.com/quickstart.md): Install the Cascade SDK and send your first trace
- [Defense Mechanism](https://docs.runcascade.com/security/adaptive-defense-mechanism.md): Detect and defend against attacks