logo
Railtracks Docs
Home
Initializing search
    railtracks
    • Home
    • Agent Development
    • AI Coding
    • Retrieval
    • Evaluations
    • Observability
    • Integrations
    • Tutorials
    • API Reference
    railtracks
    • Home
        • Quickstart
        • Installation
        • LLM Setup
        • Contributing to the Docs
        • Upgrading to 1.5.0
        • Agent Harness
        • Overview
            • Overview
            • To Do Tool
            • Key-Value Memory Tool
            • Web Search Tool
          • Function Tools
          • Agents as Tools
          • MCP
          • System Messages
          • Modalities
          • Failures
          • Common Hyperparameters
          • Structured Output
          • Overview
          • Custom
            • Overview
            • Quickstart
            • Contributing
          • Verifiers
            • Overview
            • Contributing
        • Direct Invocation
        • Flow Invocation
        • Message History
        • Context
      • Claude Code Plugin
      • AI Coding Assistants
        • Quickstart
        • Ingestion
        • Retrieval
        • Module Packages
        • Design
          • Base Class
          • Built-in Loaders
          • Base Class
          • Built-in Methods
          • Overview
          • Built-in Methods
          • Base Class
          • Backends
      • Preface
      • Quickstart
      • Visualization
        • Evaluating Agent Runs
        • Hosted Evaluations
        • Metrics Overview
        • Categorical Metrics
        • Numerical Metrics
        • Evaluator Abstraction
        • ToolUseEvaluator
        • LLMInferenceEvaluator
        • JudgeEvaluator
          • Current Visualizer
          • Legacy Visualizer
        • Cloud
        • Logging
        • Broadcasting
        • Error Handling
      • Overview
        • Providers
        • Platforms
        • Streaming
        • Github
        • Slack
        • Email
        • Overview
        • AWS S3
        • Azure Blob Storage
        • Google Cloud Storage
        • SQL Databases
        • Python Sandbox
        • Shell
        • Websearch
          • Building your first agent
          • Running your first agent
          • Agents as Tools
          • Prompts and Context
          • Flows
          • FastAPI Integration
          • Human-in-the-Loop: LLM Reviewers
          • Human-in-the-Loop: Webhook Approvals
          • Multiagent Systems
          • Multimodal Agent
          • AgentHub
          • Evaluation
        • Agent Architectures
        • FastAPI Integration
        • File Embedding
        • Retrieval Benchmark
          • Agents
          • Tools
          • MCP
          • Prompts and Context
          • RAG
          • Async/Await
          • Vector Stores
          • Overview
          • Validation Loops
          • Sequential Flows
    • API Reference
    Railtracks Logo

    Made by Agent Builders
    for Agent Builders

    Learn

    If you are curious to learn about AI Agents, how they work, and other concepts.

    Build

    If you want to build an agent harness: the loop, tools, context, and controls around a model

    Observe

    If you want to inspect, track, and visualize your Agents

    Contribute

    Something missing from the framework? Let us know!

    Next
    Quickstart