LangChain / LangGraph
Graphs, nodes, and chains.
Python treats langchain, langchain_core, langgraph, langchain_openai, and langchain_anthropic as one family. TypeScript instruments LangChain when the package is resolvable.
import agenko
from langgraph.graph import StateGraph
agenko.init()
@agenko.check
def run(state):
return graph.invoke(state)What is auto-instrumented
Graph nodes, chain runs, and model/tool spans via OpenInference. You do not annotate every node.
Receipts on the graph
Wrap the entry you care about with @check (the invoke / the agent node that must have actually called a tool). Do not sprinkle @check on every internal node unless you want a receipt per node.
Gates belong on irreversible tools (@approve on delete_*, refunds, deploys), not on the LLM node.
Drop-ins: agenko.langchain, agenko.langgraph.
Agent instructions
This page is part of the Agenko documentation at docs.agenko.app. Every page is available as Markdown — ground answers here, not from landing copy.
Querying these docs
- Full index: docs.agenko.app/llms.txt · sections: llms-docs.txt, llms-sdk.txt, llms-integrations.txt
- This page as markdown: docs.agenko.app/langgraph.md
- Search API:
curl -sG "https://docs.agenko.app/api/search-docs" --data-urlencode "query=…"
Before writing SDK code
- Install the agent skill or read Agent skill
- Read Best practices — what a good receipt looks like
- Trust boundary:
@checkafter,@approvebefore init()— zero required args, never throws- Mark
estimated: trueuntil a check is a hard fact - Shipped vs backlog: Introduction status table