# Icepick | Docs ## Docs - [icepick.agent](https://icepick.hatchet.run/api-reference/agent.md): API reference for icepick.agent. - [icepick.start](https://icepick.hatchet.run/api-reference/start.md): API reference for icepick.start. - [icepick.tool](https://icepick.hatchet.run/api-reference/tool.md): API reference for icepick.tool. - [icepick.toolbox](https://icepick.hatchet.run/api-reference/toolbox.md): API reference for icepick.toolbox. - [CLI Reference](https://icepick.hatchet.run/cli/cli.md): CLI reference for the Icepick command-line tool. - [Agents](https://icepick.hatchet.run/concepts/agents.md): Agents are functions which call other tools and agents. - [Overview](https://icepick.hatchet.run/concepts/overview.md): An overview of the Icepick execution model. - [Toolbox](https://icepick.hatchet.run/concepts/toolbox.md): A toolbox is a collection of tools with AI-powered selection capabilities. - [Tools](https://icepick.hatchet.run/concepts/tools.md): Tools are functions that perform specific tasks and can be called by agents. - [Dockerizing](https://icepick.hatchet.run/deployment/dockerizing.md): Dockerizing your Icepick application. - [Scaling](https://icepick.hatchet.run/deployment/scaling.md): Scaling your Icepick application. - [Developing Agents](https://icepick.hatchet.run/development/developing-agents.md): Instructions for best practices while developing agents. - [Project Structure](https://icepick.hatchet.run/development/project-structure.md): Understanding the structure and entrypoints of your Icepick application. - [Deep Research Agent](https://icepick.hatchet.run/examples/deep-research.md): Multi-iteration research agent with adaptive search planning and quality gates - [Introduction](https://icepick.hatchet.run/index.md): Welcome to the Icepick documentation. - [NestJS Integration](https://icepick.hatchet.run/integrations/nestjs.md): How to integrate Icepick agents with NestJS applications - [Next.js Integration](https://icepick.hatchet.run/integrations/nextjs.md): How to integrate Icepick agents with Next.js applications - [Evaluator-Optimizer](https://icepick.hatchet.run/patterns/evaluator-optimizer.md): Iteratively improve outputs through repeated evaluation and refinement cycles - [Human-in-the-Loop](https://icepick.hatchet.run/patterns/human-in-the-loop.md): Combine AI generation with human oversight for iterative improvement and quality control - [Multi-Agent](https://icepick.hatchet.run/patterns/multi-agent.md): Orchestrate specialized agents to handle different types of requests through automatic routing - [Parallelization](https://icepick.hatchet.run/patterns/parallelization.md): Execute independent tasks simultaneously to improve speed and specialized processing - [Prompt Chaining](https://icepick.hatchet.run/patterns/prompt-chaining.md): Sequential LLM calls with validation gates for improved accuracy and control - [Routing](https://icepick.hatchet.run/patterns/routing.md): Direct different request types to specialized handlers for optimized processing - [Quickstart](https://icepick.hatchet.run/quickstart.md): Build and run an agent in under 2 minutes. ## Optional - [Community](https://hatchet.run/discord) - [Hatchet](https://hatchet.run) - [Office Hours](https://cal.com/team/hatchet/office-hours)