# When I call a MCP Tool using FastMCP, I am not getting Context initialized correctly
**URL:** https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354
**Category:** LangChain
**Tags:** python-help
**Created:** [April 9, 2026, 1:36pm UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354 "2026-04-09T13:36:46Z")
**Posts on this page:** 6
**Page:** 1
### Author:  [@tuhin-52321](https://forum.langchain.com/u/tuhin-52321)
#### Post date: [April 9, 2026, 1:36pm UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354/1 "2026-04-09T13:36:46Z")
Langchain version: 1.2.15
langchain-mcp-adapters version: 0.2.2
FastMCP Version : 3.2.0
My MCP tool needs access to ctx.transport value, (stdio, sse etc.) to do some specific processing. Sample tool as below:
@tally\_mcp.tool()
async def my\_tool(ctx: Context) → dict:
```
"""My tool access data based on transport."""
transport=ctx.transport #introduced in FastMCP v3
return { "transport": transport }
```
At client side my langchain code is as below:
```python
client = MultiServerMCPClient(
{
"mcp_server": {
"transport": "stdio",
"command": "python",
"args": ["mcp_stdio_server.py"]
}
}
)
tools = await client.get_tools()
agent = create_agent(
model=…,
tools=tools
)
response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "list my companies."}]}
)
```
```
In the tool response I am getting:
```
[{‘type’: ‘text’, ‘text’: ‘{ **“transport”: null** \n}’, ‘id’: ‘lc\_2e6e7b99-d184-4b40-93f8-3acabdeffe33’}]
This should be set as ‘stdio’. is it a bug in Langchain?
---
### Author:  [@keenborder786](https://forum.langchain.com/u/keenborder786)
#### Post date: [April 9, 2026, 4:05pm UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354/2 "2026-04-09T16:05:52Z")
Hello @tuhin-52321
I replicated your flow with following versions:
```auto
langchain-mcp-adapters: 0.2.2
fastmcp: 3.2.2
langchain: 1.2.15
```
## server.py
```python
"""FastMCP server demonstrating ctx.transport access in a tool."""
from fastmcp import FastMCP, Context
tally_mcp = FastMCP(name="tally-mcp-server")
@tally_mcp.tool()
async def my_tool(ctx: Context) -> dict:
"""Return transport info and list sample companies based on the transport type."""
transport = ctx.transport
companies = ["Acme Corp", "Globex Inc", "Initech Ltd"]
return {
"transport": transport,
"companies": companies,
"note": f"Fetched via {transport!r} transport",
}
if __name__ == " __main__":
tally_mcp.run(transport="stdio")
```
## client.py
```python
async def main() -> None:
client = MultiServerMCPClient(
{
"mcp_server": {
"transport": "stdio",
"command": "python3",
"args": ["mcp_stdio_server.py"],
}
}
)
tools = await client.get_tools()
print(f"Loaded {len(tools)} tool(s): {[t.name for t in tools]}")
model = ChatOpenAI(model="gpt-4o-mini")
agent = create_agent(model=model, tools=tools)
response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "list my companies."}]}
)
for msg in response["messages"]:
role = getattr(msg, "type", msg. __class__. __name__ )
content = msg.content
if content:
print(f"\n[{role}]: {content}")
if __name__ == " __main__":
asyncio.run(main())
```
And got the following output:
```bash
[04/09/26 20:59:32] INFO Starting MCP server 'tally-mcp-server' with transport 'stdio' transport.py:209
Loaded 1 tool(s): ['my_tool']
╭──────────────────────────────────────────────────────────────────────────────╮
│ │
│ │
│ ▄▀▀ ▄▀█ █▀▀ ▀█▀ █▀▄▀█ █▀▀ █▀█ │
│ █▀ █▀█ ▄▄█ █ █ ▀ █ █▄▄ █▀▀ │
│ │
│ │
│ FastMCP 3.2.2 │
│ https://gofastmcp.com │
│ │
│ 🖥 Server: tally-mcp-server, 3.2.2 │
│ 🚀 Deploy free: https://horizon.prefect.io │
│ │
╰──────────────────────────────────────────────────────────────────────────────╯
[04/09/26 20:59:36] INFO Starting MCP server 'tally-mcp-server' with transport 'stdio' transport.py:209
[human]: list my companies.
[tool]: [{'type': 'text', 'text': '{"transport":"stdio","companies":["Acme Corp","Globex Inc","Initech Ltd"],"note":"Fetched via \'stdio\' transport"}', 'id': 'lc_2cd0cb69-1b5b-46ed-9b21-503b6fdc3f17'}]
[ai]: Here are your companies:
1. Acme Corp
2. Globex Inc
3. Initech Ltd
```
As you can see, I was able to access the transport attribute and did not get a null, which you are getting.
The only difference I see is that I am using the latest fast mcp version, so perhaps try updating it.
---
### Author:  [@tuhin-52321](https://forum.langchain.com/u/tuhin-52321)
#### Post date: [April 10, 2026, 4:00am UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354/3 "2026-04-10T04:00:40Z")
I upgraded to FastMCP 3.2.3:
Name: fastmcp
Version: 3.2.3
Summary: The fast, Pythonic way to build MCP servers and clients.
Home-page: [https://gofastmcp.com](https://gofastmcp.com)
Author: Jeremiah Lowin
Author-email:
License-Expression: Apache-2.0
Location: d:\work\code.tallyai.backend.venv\Lib\site-packages
Requires: authlib, cyclopts, exceptiongroup, httpx, jsonref, jsonschema-path, mcp, openapi-pydantic, opentelemetry-api, packaging, platformdirs, py-key-value-aio, pydantic, pyperclip, python-dotenv, pyyaml, rich, uncalled-for, uvicorn, watchfiles, websockets
Required-by:
(.venv) PS D:\work\code.tallyai.backen
Still no luck.
AI response:
Tool response: [{‘type’: ‘text’, ‘text’: ‘{\n “transport”: null,\n “companies”: [\n “Acme Corp”,\n “Globex Inc”,\n “Initech Ltd”\n],\n “note”: “Fetched via None transport”\n}’, ‘id’: ‘lc\_0541a012-cd29-4337-acb8-bad716058f2e’}]
AI response: Here are your companies: Acme Corp, Globex Inc, Initech Ltd.
Another difference, I am using **ChatGoogleGenerativeAI(vertexai=True,** …) , not **ChatOpenAI.**
I am using Gemini 2.5 Flash via Vertex AI. Could that be the issue?
I am using Windows OS for testing - can this be cause of the issue?
UPDATE:
Found the problem - it is Windows OS issue. Same code working fine in my Mac. With Mac the output coming correctly. But with Windows , it is not working. In Windows I am not even getting the info log when mc server starts in stdio mode “ [04/09/26 20:59:36] INFO Starting MCP server ‘tally-mcp-server’ with transport ‘stdio’“
Is FastMCP not supported in Windows?
---
### Author:  [@tuhin-52321](https://forum.langchain.com/u/tuhin-52321)
#### Post date: [April 10, 2026, 10:35am UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354/4 "2026-04-10T10:35:32Z")
UPDATE:
Created below setup:
1. MCP Server running on Docker
2. Langchain is running on Windows.
Still same issue. No idea what is happening. This time I used HTTP Streamble transport.
Docker run output:
INFO: Started server process [7]
INFO: Waiting for application startup.
INFO | mcp.server.streamable\_http\_manager | streamable\_http\_manager.py:128 | StreamableHTTP session manager started
INFO: Application startup complete.
INFO: Uvicorn running on [http://0.0.0.0:8000](http://0.0.0.0:8000) (Press CTRL+C to quit)
INFO | mcp.server.streamable\_http\_manager | streamable\_http\_manager.py:255 | Created new transport with session ID: fa3cd8fb3c2941f1b455a9b90ac00ba6
INFO: 172.17.0.1:41112 - “POST /mcp HTTP/1.1” 200 OK
INFO: 172.17.0.1:41122 - “POST /mcp HTTP/1.1” 202 Accepted
INFO: 172.17.0.1:41126 - “GET /mcp HTTP/1.1” 200 OK
INFO: 172.17.0.1:41136 - “POST /mcp HTTP/1.1” 200 OK
INFO | mcp.server.lowlevel.server | server.py:727 | Processing request of type ListToolsRequest
INFO | mcp.server.streamable\_http | streamable\_http.py:785 | Terminating session: fa3cd8fb3c2941f1b455a9b90ac00ba6
INFO: 172.17.0.1:41150 - “DELETE /mcp HTTP/1.1” 200 OK
INFO | mcp.server.streamable\_http\_manager | streamable\_http\_manager.py:255 | Created new transport with session ID: 5f5d5c8550f94b5f8d9fcd9cb85b233c
INFO: 172.17.0.1:43074 - “POST /mcp HTTP/1.1” 200 OK
INFO: 172.17.0.1:43076 - “POST /mcp HTTP/1.1” 202 Accepted
INFO: 172.17.0.1:43084 - “GET /mcp HTTP/1.1” 200 OK
INFO: 172.17.0.1:43086 - “POST /mcp HTTP/1.1” 200 OK
INFO | mcp.server.lowlevel.server | server.py:727 | Processing request of type CallToolRequest
INFO | mcp\_tools\_logger | tally\_mcp\_tools.py:19 | [1] Received request with transport: None
INFO: 172.17.0.1:43102 - “POST /mcp HTTP/1.1” 200 OK
INFO | mcp.server.lowlevel.server | server.py:727 | Processing request of type ListToolsRequest
INFO | mcp.server.streamable\_http | streamable\_http.py:785 | Terminating session: 5f5d5c8550f94b5f8d9fcd9cb85b233c
INFO: 172.17.0.1:43116 - “DELETE /mcp HTTP/1.1” 200 OK
MCP Client output:
(.venv) PS D:\work\code.tallyai.backend\conversational-ai\infra\cdk\lib\docker\llm-ws\> python .\tally\_mcp\_client\_test.py
[human]: list my companies.
[tool]: [{‘type’: ‘text’, ‘text’: ‘{\n “transport”: null,\n “companies”: [\n “Acme Corp”,\n “Globex Inc”,\n “Initech Ltd”\n],\n “note”: “Fetched via None transport”\n}’, ‘id’: ‘lc\_619d944a-f402-4865-9ef6-569ed76ae5d2’}]
[ai]: Here are your companies: Acme Corp, Globex Inc, Initech Ltd.
(.venv) PS D:\work\code.tallyai.backend\conversational-ai\infra\cdk\lib\docker\llm-ws\>
is the issue in client side of Windows?
**Working when client is run from Mac, but not working when client is run from Windows.**
Client Code:
tally\_mco\_client\_test.py
```python
import os
os.environ["MCP_PORT"] = "8000"
import asyncio
from langchain.agents import create_agent
mcp_port = os.getenv("MCP_PORT", "8000")
def get_mcp_client(thread_id: str, company_id: str, mobile_number: str) -> MultiServerMCPClient:
tally_mcp_client = MultiServerMCPClient(
{
"tally_mcp_server": {
"transport": "streamable_http",
"url": "http://localhost:" + mcp_port + "/mcp"
}
}
)
return tally_mcp_client
async def main():
client = get_mcp_client(thread_id, company_id, mobile_number)
#Get Tools
tools = await client.get_tools()
agent = create_agent(
model=...,
tools=tools
)
tally_response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "list my companies."}]}
)
for msg in tally_response["messages"]:
role = getattr(msg, "type", msg. __class__. __name__ )
content = msg.content
if content:
print(f"\n[{role}]: {content}")
if __name__ == " __main__":
asyncio.run(main())
```
Docker Directory For MCP Server:
Dockerfile:
```auto
# Step 1: Use the official Python image (LTS version)
FROM python:3.13
ARG MCP_PORT
ARG LOG_LEVEL
# Step 2: Set the working directory inside the container
WORKDIR /usr/src/app
COPY requirements.txt ./
COPY *.py ./
# Step 3 : Install required python packages
RUN pip install --upgrade pip
RUN pip install -r requirements.txt
# Step 4: Expose the port your application will run on (e.g., 80)
# EXPOSE Ports
EXPOSE ${MCP_PORT}
#Environment Variables
ENV MCP_PORT=${MCP_PORT}
ENV LOG_LEVEL=${LOG_LEVEL}
# Step 6: Set the command to start the application
COPY --chmod=755 < dict:
"""Return transport info and list sample companies based on the transport type."""
transport = ctx.transport
companies = ["Acme Corp", "Globex Inc", "Initech Ltd"]
return {
"transport": transport,
"companies": companies,
"note": f"Fetched via {transport!r} transport",
}
```
My client side requirement.txt file (is this a problem?)
```auto
langchain-google-vertexai~=3.2.2
boto3~=1.42.83
fastapi~=0.124.4
pandas~=2.3.3
psutil~=7.1.3
langchain-tavily~=0.2.17
cryptography~=46.0.6
langchain-community~=0.4.1
valkey~=6.1.1
langchain_aws~=1.1.0
scipy~=1.16.3
langchain_chroma~=1.0.0
langchain_milvus~=0.3.3
tiktoken~=0.12.0
jellyfish~=1.2.1
langgraph-checkpoint-postgres~=3.0.5
uvicorn-worker~=0.4.0
uvicorn~=0.38.0
gunicorn~=23.0.0
aiohttp~=3.13.5
google-auth~=2.49.1
google-auth-oauthlib~=1.3.1
langchain~=1.2.15
langchain-core~=1.2.26
langchain-google-genai~=4.2.1
google-genai~=1.66.0
google-cloud-aiplatform~=1.140.0
vertexai~=1.43.0
langgraph~=1.1.6
langgraph-supervisor~=0.0.31
langgraph-prebuilt~=1.0.9
fastmcp~=3.2.2
langchain-mcp-adapters~=0.2.2
```
---
### Author:  [@keenborder786](https://forum.langchain.com/u/keenborder786)
#### Post date: [April 10, 2026, 7:48pm UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354/5 "2026-04-10T19:48:40Z")
Since I am not able to replicate this issue on Linux, I believe this may be a bug in FastMCP rather than in the LangChain MCP adapter, as the `transport` attribute in the tool is isolated to the FastMCP package.
I think you should open an issue on the official FastMCP repository. I also checked online and found no similar reported issues.
Additionally, since the `transport` attribute was introduced very recently in FastMCP, there is a good chance there is a bug specific to the Windows version.
---
### Author:  [@tuhin-52321](https://forum.langchain.com/u/tuhin-52321)
#### Post date: [April 13, 2026, 5:54am UTC](https://forum.langchain.com/t/when-i-call-a-mcp-tool-using-fastmcp-i-am-not-getting-context-initialized-correctly/3354/6 "2026-04-13T05:54:27Z")
> [@tuhin-52321](#):
>
> ` mcp.server.fastmcp`
Found the issue - it was issue with import …
import should be “from fastmcp import FastMCP”
This solved the issue.