Esta página se muestra en inglés. Una traducción al español está en curso.
KnowledgeMay 14, 2026·2 min de lectura

Claude Cookbooks — Claude API Notebook Recipes

Claude Cookbooks is a repo of copyable notebooks for Claude API patterns (tool use, RAG, vision, evals) so you can prototype faster (42.9k★).

Agent Toolkit
Agent Toolkit · Community
Listo para agents

Este activo puede ser leído e instalado directamente por agents

TokRepo expone un comando CLI universal, contrato de instalación, metadata JSON, plan según adaptador y contenido raw para que los agents evalúen compatibilidad, riesgo y próximos pasos.

Needs Confirmation · 64/100Política: confirmar
Superficie agent
Cualquier agent MCP/CLI
Tipo
Knowledge
Instalación
Single
Confianza
Confianza: Established
Entrada
Asset
Comando CLI universal
npx tokrepo install 25da083c-f4af-52f0-8e34-284ac1d865de
contrato de instalaciónJSON de metadataplan adaptadorcontenido raw
Introducción

Claude Cookbooks is a repo of copyable notebooks for Claude API patterns (tool use, RAG, vision, evals) so you can prototype faster (42.9k★).

Best for: Developers who want vetted Claude API patterns they can copy into apps (tool use, RAG, multimodal, evaluations)

Works with: Python + Jupyter (for notebooks) and any language that can call the Claude API; start from the recipe folders and adapt

Setup time: 5-15 minutes

Main

  • Recipe-first structure: the README links to focused folders (classification, RAG, summarization, tool use) so you can start from a minimal working example.
  • Tool use examples: includes practical notebooks for function/tool integration (e.g., calculator, customer service patterns).
  • Multimodal coverage: contains vision and PDF-related recipes so you can wire image/PDF inputs into your app workflows.
  • Evaluation recipes: ships notebooks for building eval loops—useful when you need a repeatable quality gate for prompts or tools.

FAQ

  • Do I need to use notebooks?: No. Use notebooks as a reference and copy the minimal code into your own runtime.
  • Where should I start?: Start from one capability folder (tool use or RAG) and run a single recipe end to end before mixing patterns.
  • Is it Python-only?: Most examples are in Python notebooks, but the patterns (prompting, tool use, evals) translate to any language.
🙏

Fuente y agradecimientos

Source: https://github.com/anthropics/claude-cookbooks > License: MIT > GitHub stars: 42907 · forks: 4841

Discusión

Inicia sesión para unirte a la discusión.
Aún no hay comentarios. Sé el primero en compartir tus ideas.

Activos relacionados

Awesome Claude DXT — Desktop Extensions Directory

Directory of Claude Desktop extensions (DXT). Discover installable desktop integrations and learn patterns beyond MCP servers.

Knowledge
AI Open Source

Claude Howto — Guided Claude Code Learning Path

Claude Howto is a visual, example-driven roadmap for Claude Code that ships copy-paste `.claude/` templates, quizzes, and a 15-minute quick start.

Knowledge
Prompt Lab

auto-memory — Keep CLAUDE.md Always in Sync

Claude Code plugin that keeps CLAUDE.md in sync with file changes; verified 143★ with marketplace install and `/auto-memory:*` commands.

Knowledge
Prompt Lab

MemoryGraph — Graph MCP Memory Server (pipx)

MemoryGraph is a graph-based MCP memory server for coding agents; verified 199★ and shows a pipx install + `claude mcp add` setup for Claude Code.

Knowledge
Agent Toolkit
◈Inicio🔍Buscar👤Yo