- Home
- MCP servers
- SEC Filing
SEC Filing
- python
0
GitHub Stars
python
Language
7 months ago
First Indexed
3 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"sharhadbashar-sec-filing-mcp": {
"command": "/Users/sharhad/.pyenv/shims/python3",
"args": [
"/Users/sharhad/mcp/server/server.py"
]
}
}
}MCP is a modular platform that combines a server, a client, and an ingest component to store and query data (such as SEC filings) using Pinecone. You run a local server for tools like Claude Desktop or host it remotely for a web frontend, and you interact with it through a Streamlit client or other MCP-enabled clients.
How to use
To use the MCP server with a client, start the local server or a hosted server, then connect your client application to that server. The local server lets you run entirely on your machine, while a hosted server enables access from a web or desktop frontend.
Local server (Claude Desktop) setup: you run the MCP server as a standalone Python process and point Claude Desktop at it via a configuration entry. The server exposes the endpoints and tools you’ll use from the client.
Hosted server (front-end client) setup: you run the MCP server in a background process and run a front-end client (such as a Streamlit app) that talks to the server. The client provides an interface to query and retrieve documents, with re-ranking and document retrieval handled by the server.
How to install
Prerequisites you need before installing: Python 3.x installed on your system.
Run the local server used by Claude Desktop with the following steps.
{
"mcpServers": {
"sec-filing-server": {
"command": "/Users/sharhad/.pyenv/shims/python3", # Path to python executable
"args": ["/Users/sharhad/mcp/server/server.py"] # Absolute path to server.py
}
}
}
Additional sections
You can also run a hosted MCP server and connect a frontend client. The hosted server can be started from the server folder and the client from the client folder.
Example steps shown for running the hosted setup:
# Start the hosted MCP server
python server_host.py
# In a separate terminal, start the Streamlit client
cd client
streamlit run app.py
Notes and examples
The MCP project includes data ingest and two core tools. You enable access to the functions via the server and then use them from the client.
Available tools described in the system are:
-
get_sec_files: Takes a company ticker and returns the SEC filing files available for that company.
-
query_sec: Takes your query and returns an answer based on the SEC filings and their summaries.
Available tools
get_sec_files
Retrieves the list of SEC filing files available for a given company ticker from the MCP server.
query_sec
Answers user questions by querying the SEC filings and their summaries via the MCP server.