diff --git a/README.md b/README.md
index da0f2fd..12efbeb 100644
--- a/README.md
+++ b/README.md
@@ -6,6 +6,26 @@ Convert CLI-based AI agents (Claude Code, etc.) to OpenAI-compatible API endpoin
 
 This adapter allows you to use local CLI tools like Claude Code as drop-in replacements for OpenAI's API in your development environment, while keeping the same code structure for production.
 
+LangChain Demos
+- JavaScript/TypeScript (Node): see `examples/langchain-js`. Minimal usage:
+  ```ts
+  import { ChatOpenAI } from "@langchain/openai";
+  const llm = new ChatOpenAI({
+    apiKey: "dummy-key",
+    baseURL: "http://localhost:8000/v1",
+    model: "claude-code",
+  });
+  const resp = await llm.invoke("hi!");
+  console.log(resp.content);
+  ```
+
+- Python: see `examples/langchain-py`. Minimal usage:
+  ```py
+  from langchain_openai import ChatOpenAI
+  llm = ChatOpenAI(api_key="dummy-key", base_url="http://localhost:8000/v1", model="claude-code")
+  print(llm.invoke("hi!").content)
+  ```
+
 **Use Cases:**
 - **Production**: Use OpenAI API (pay per token)
 - **Development**: Use local Claude Code with Haiku model (reduce costs)
diff --git a/examples/langchain-js/README.md b/examples/langchain-js/README.md
new file mode 100644
index 0000000..4577ccd
--- /dev/null
+++ b/examples/langchain-js/README.md
@@ -0,0 +1,23 @@
+LangChain JS Demo (OpenAI-compatible Adapter)
+
+This example shows how to call the local adapter using LangChain JS.
+
+Prerequisites
+- Adapter server running locally: `npm run dev` (defaults to http://localhost:8000)
+- Node.js >= 20
+
+Setup
+```
+cd examples/langchain-js
+npm init -y
+npm i @langchain/openai langchain
+```
+
+Run
+```
+node chat.mjs
+```
+
+Files
+- `chat.mjs`: Minimal chat completion using `ChatOpenAI` with a custom `baseURL`.
+
diff --git a/examples/langchain-js/chat.mjs b/examples/langchain-js/chat.mjs
new file mode 100644
index 0000000..ffcf50b
--- /dev/null
+++ b/examples/langchain-js/chat.mjs
@@ -0,0 +1,12 @@
+import { ChatOpenAI } from "@langchain/openai";
+
+// Point LangChain to the local adapter's OpenAI-compatible endpoint
+const llm = new ChatOpenAI({
+  apiKey: process.env.OPENAI_API_KEY || "dummy-key", // not used by local adapter
+  baseURL: process.env.ADAPTER_BASE_URL || "http://localhost:8000/v1",
+  model: process.env.ADAPTER_MODEL || "claude-code",
+});
+
+const resp = await llm.invoke("hi! Reply in one short sentence.");
+console.log(resp.content);
+
diff --git a/examples/langchain-py/README.md b/examples/langchain-py/README.md
new file mode 100644
index 0000000..15fe0e8
--- /dev/null
+++ b/examples/langchain-py/README.md
@@ -0,0 +1,24 @@
+LangChain Python Demo (OpenAI-compatible Adapter)
+
+This example shows how to call the local adapter using LangChain for Python.
+
+Prerequisites
+- Adapter server running locally: `npm run dev` (defaults to http://localhost:8000)
+- Python 3.9+
+
+Setup
+```
+cd examples/langchain-py
+python -m venv .venv
+source .venv/bin/activate  # Windows: .venv\\Scripts\\activate
+pip install langchain langchain-openai
+```
+
+Run
+```
+python demo.py
+```
+
+Files
+- `demo.py`: Minimal chat completion using `ChatOpenAI` with a custom `base_url`.
+
diff --git a/examples/langchain-py/demo.py b/examples/langchain-py/demo.py
new file mode 100644
index 0000000..79d5c4f
--- /dev/null
+++ b/examples/langchain-py/demo.py
@@ -0,0 +1,12 @@
+from langchain_openai import ChatOpenAI
+
+# Point LangChain to the local adapter's OpenAI-compatible endpoint
+llm = ChatOpenAI(
+    api_key="dummy-key",  # not used by local adapter
+    base_url="http://localhost:8000/v1",
+    model="claude-code",
+)
+
+resp = llm.invoke("hi! Reply in one short sentence.")
+print(resp.content)
+