LangChain cogDepot Prompt/cogdepot_plan_my_spend
How to: LangChain cogDepot cogdepot_plan_my_spend
Add cogDepot 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({
"cogdepot": {
"transport": "streamable_http",
"url": "https://mcp.cogdepot.com",
}
})
tools = await client.get_tools()
-
02
Save the file and restart LangChain.
-
03
cogdepot_plan_my_spend is surfaced through the client's prompt API, fetched by name. Unlike a tool, you invoke it deliberately — LangChain will not call it for you.
-
04
If it does not appear, check that cogDepot is connected and that you are looking at its prompts rather than its tools.
get_tools() returns the server's tools as LangChain tools, ready for an agent. LangChain docs
Set these first
cogDepot will not start until these are set, so it never reaches LangChain.
-
COGDEPOT_API_KEY -
COGDEPOT_API_BASE_URL -
COGDEPOT_OAUTH_ISSUER -
COGDEPOT_OAUTH_CLIENT_ID -
COGDEPOT_OAUTH_RESOURCE -
COGDEPOT_OAUTH_SCOPES
Same prompt, other clients
Read from the server itself, by connecting to it and calling prompts/list on 24 September 2026.
A listing does not declare its prompts, so asking is the only way to know them, and this
is what the server actually offers rather than what its listing claims. Names only are
stored; connect the server for each prompt's arguments.