LangChain TokenLab MCP Server open_tokenlab_model_explorer
How to: LangChain TokenLab MCP Server open_tokenlab_model_explorer
Add TokenLab MCP Server to LangChain
-
01
Open pip install langchain-mcp-adapters and merge this in. Keep any servers already there.
from langchain_mcp_adapters.client import MultiServerMCPClient
client = MultiServerMCPClient({
"tokenlab": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "@tokenlabai/mcp-server"],
}
})
tools = await client.get_tools()
-
02
Save the file and restart LangChain.
-
03
Ask for something open_tokenlab_model_explorer does. LangChain lists the server's tools on connect and calls open_tokenlab_model_explorer itself — you do not invoke it by name.
-
04
If nothing happens, check the server is running and that TokenLab MCP Server's identifier in your config matches the one above exactly.
get_tools() returns the server's tools as LangChain tools, ready for an agent. LangChain docs
Set these first
TokenLab MCP Server will not start until these are set, so open_tokenlab_model_explorer never becomes available.
-
TOKENLAB_API_BASE -
TOKENLAB_API_KEY -
TOKENLAB_REQUEST_TIMEOUT_MS -
TOKENLAB_MCP_TOOL_PROFILE -
TOKENLAB_MCP_SCHEMA_MODE -
TOKENLAB_MCP_MAX_FILE_BYTES -
TOKENLAB_MCP_INLINE_BYTES -
TOKENLAB_ARTIFACT_DIR -
TOKENLAB_MANAGEMENT_TOKEN
Same tool, other clients
This list was read from the server itself, by connecting to it and calling tools/list on 18 October 2026. It is what
the server actually exposes, not what its listing claims.
Mutating and Read-only are read off each tool's name, not its schema — a hint, not a guarantee. The registry stores tool names only; connect the server for its live schemas.