- DATA
- ANALYSIS
- REPORTS
- HUMAN
Short answer
How can I pull specific financial data from SEC filings using AI?
Your AI uses the SEC EDGAR connector to query XBRL data points directly from company filings. You can ask for specific metrics like net income or total assets, and your agent will fetch the exact values you need. Here is how the data moves from the SEC to your chat.
The results.
Where the data lands.
The AI processes the raw SEC data into structured information you can actually use.
DATA
Structured XBRL values
The AI pulls specific numbers like assets or revenue. You get clean data instead of reading through massive text documents.
ANALYSIS
Comparative insights
Once the facts are pulled, your agent can compare figures across different quarters or years. This makes spotting trends much easier.
REPORTS
Drafted summaries
Your agent can take the extracted facts and write up a summary. This turns raw SEC data into a readable format for your notes.
HUMAN
Final verification
You stay in control. You review the numbers the AI pulled to ensure they match the filing before you use them in your work.
The workflow
What your AI does when the request arrives.
The AI acts as a bridge between your chat interface and the SEC's structured data.
Identify the concept
The AI first determines the exact XBRL tag needed for the financial metric you asked about.
get_company_conceptFetch company data
The agent queries the SEC database to find the specific facts associated with that company's identifier.
get_company_factsFilter and extract
The AI narrows down the massive list of available facts to find the specific values you requested.
get_company_factsFormat the answer
The agent translates the raw XBRL data into a plain English response for your chat window.
get_company_concept
Try it
Copy these to start.
Use these prompts to begin your research immediately.
Starting points
Swap out the company names and specific metrics for the ones you are researching.
SEC EDGAR Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_1OvhyNghoLef4hJMsgJc7Zjxy1eriC0rizTeQFM5/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — SEC EDGAR capabilities are ready to use.
{
"mcpServers": {
"sec-edgar-mcp": {
"url": "https://edge.vinkius.com/vk_preview_1OvhyNghoLef4hJMsgJc7Zjxy1eriC0rizTeQFM5/mcp"
}
}
}What was Apple's net income in their last 10-K filing?
Find the total assets for Microsoft from their most recent quarterly report.
Compare the revenue for Tesla across the last three fiscal quarters.
List the operating expenses for Amazon as reported in their latest filing.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect SEC EDGAR once, then ask.
One click connects the SEC EDGAR MCP to your AI client. Your credentials stay encrypted, and you can start querying filings immediately.
Connect SEC EDGAR to your AIFAQ
How this task behaves.
- 01
Can the AI change the data in the SEC filings?
No. The AI only reads data from the SEC EDGAR database. It cannot modify or delete any official filings.
- 02
Does the AI perform the final audit of the numbers?
No. The AI extracts the data, but you are responsible for verifying the numbers against the original filings for accuracy.
- 03
Can I use this to get data from non-SEC filings?
No. This connector is specifically built to interface with the SEC EDGAR database.
- 04
What happens if a company hasn't filed a specific metric?
If the requested XBRL concept does not exist in the company's filings, the AI will tell you it cannot find that data.
- 05
Can the AI download the entire filing as a PDF?
No. The tools are designed to extract specific XBRL data points, not to download or generate document files.
More questions about SEC EDGAR? The Connector page answers them. See everything the SEC EDGAR Connector can do
Connect SEC EDGAR to Claude, Cursor, ChatGPT & more
