Cheatsheets
Dense, scannable one-pagers for the tools you already know — the commands, classes, and patterns worth keeping under Ctrl + F. Not tutorials, not deep dives. Just the parts you forget.
LangChain
LangChain is a Python/JS framework for composing LLM-powered applications: it provides a unified interface across model providers, a composable LCEL pipe syntax, and pre-built chains for RAG, agents, and memory. This cheatsheet covers the core imports, Chat and Embedding model wrappers, LCEL runnables, prompt templates, retrievers, agents, tools, LangGraph state, and full minimal RAG and agent examples.
CrewAI
Agents, Tasks, Crews, Processes, Flows — the everyday API surface plus the YAML config that drives a crew.
Model Context Protocol
JSON-RPC primitives, tools, resources, prompts, sampling, lifecycle — the spec compressed into a one-page reference.
Prompt Engineering
Zero-shot, few-shot, chain-of-thought, role prompting, structured output, tool use — patterns and gotchas in one place.
Git
Branching, rebasing, fixing mistakes, recovering lost commits, interactive rebase, worktrees — the commands you reach for under pressure.
Docker
Images, containers, volumes, networks, Compose, multi-stage builds, and BuildKit — the day-to-day Docker commands and Compose snippets for Docker Engine 25+ and Compose v2. Covers the full workflow from Dockerfile authoring through registry push and production cleanup.
Regex
Character classes, anchors, quantifiers, groups, lookarounds — with examples that actually parse the things you need.
Python
Comprehensions, dataclasses, typing, async, pathlib, itertools — the standard-library surface a Python dev actually uses.
LlamaIndex
LlamaIndex is a Python data framework for building LLM applications over custom documents: it handles ingestion (SimpleDirectoryReader, PDF, web), chunking, embedding, vector storage, and retrieval in a single opinionated pipeline. This cheatsheet covers Documents, Nodes, VectorStoreIndex, retrievers, query engines, the chat engine pattern, agents, tools, and production configuration.
Haystack
Components, pipelines, document stores, retrievers, generators, agents — the deepset framework as a one-page reference.
AutoGen
AgentChat, Core, Extensions — agents, teams, termination, tools, and the modern v0.4+ async API. Covers AssistantAgent, RoundRobinGroupChat, SelectorGroupChat, and the autogen_ext model clients for OpenAI, Anthropic, and Ollama.
LangSmith
LangSmith is LangChain's observability and evaluation platform: every chain and agent run is traced automatically, letting you inspect each step's inputs, outputs, latency, and token usage. This cheatsheet covers tracing setup, the Client API, dataset creation, evaluators, feedback logging, the Prompt Hub, and automated testing patterns for LLM applications in production.