mirror of
https://github.com/magnus919/agent-skills.git
synced 2026-09-12 12:06:29 +03:00
Magnus Hedemark
fe3be63800
Greenfield SkillOpt: 3 epochs optimizing discoverability, decision guidance, and troubleshooting for a brand-new LangChain skill. Epoch 1 — Prominence: - Added critical AgentExecutor deprecation callout at top - Framework Routing Guide for cross-portfolio decisions Epoch 2 — Decision Guidance: - Where to Start table with AgentExecutor migration row - Pipeline Mode table (Quick/RAG/Agent/Production) Epoch 3 — Pattern Expansion: - Troubleshooting table with reference file links - FAQ section covering installation, migration, performance 13 files: SKILL.md, 7 references, 4 templates, 1 script. v1.0.0 -> v1.0.3 across 3 SkillOpt epochs. Signed-off-by: Magnus Hedemark <magnus919@pm.me>
1.3 KiB
1.3 KiB
LangChain Architecture
Package Structure
| Package | Purpose |
|---|---|
langchain-core |
Base abstractions: Runnable, prompts, messages, LLMs, tools |
langchain |
Meta-package with prebuilt chains, agents, retrieval |
langchain-community |
Third-party integrations (optional deps) |
langchain-openai |
OpenAI/ChatOpenAI wrapper |
langchain-anthropic |
Anthropic Claude wrapper |
langchain-google |
Google Gemini wrapper |
langchain-mcp-adapters |
MCP server tool integration |
The Runnable Protocol
Every LangChain component implements Runnable, a standardized protocol enabling:
chain = prompt | model | parser
All Runnables support invoke, ainvoke, stream, batch, and astream_events.
v1.0 Changes (October 2025)
- Agents now run on LangGraph runtime via
create_agent - AgentExecutor in maintenance mode until Dec 2026
- LCEL is the sole recommended chain composition method
- Legacy
LLMChainfully deprecated - Enhanced streaming API with event types
Installation
pip install langchain langchain-openai python-dotenv
# For specific integrations:
pip install langchain-anthropic langchain-google
pip install langchain-community
pip install langchain-mcp-adapters
Requires Python 3.10+. Python 3.11+ recommended.