Pull quarterly filing metrics for stock valuation

A developer used ChatGPT Agent to gather key figures from SEC 10-Qs for four stocks and feed a valuation workflow.

Done with ChatGPT Agent· Intermediate

What they did
They asked ChatGPT Agent mode, relying on its web search, to read the latest public 10-Q filings for AAPL, MSFT, NVDA, and TSLA. The agent produced a set of fundamentals for each ticker, including the figure, whether the metric was moving up or down, a 1-5 score, a short explanation, the filing link, and the arithmetic behind the number. Claude helped draft the prompts, and the results were passed into a Fundamental agent in a LangGraph pipeline.
How it went
The pipeline used those inputs to combine valuation, sentiment, and fundamental votes, then backtested the picks over 63 trading days. Building the prototype took about 8-10 hours. The README did not give an accuracy score for the extracted values.
Tips
Restrict the filings to documents that were public before your decision date so you do not leak future information into the workflow. Ask for both the source link and the calculation steps for each metric so the output is easier to check.

Try it yourself with ChatGPT Agent

For [ticker symbols], find the newest SEC 10-Q that was available by [decision date]. For each company, report year-over-year revenue change, free cash flow, and net cash or net debt. Put the findings in a table and include: the number, the direction of change, a 1-5 score, a brief explanation in plain language, the SEC URL, and the exact calculation used. If a figure cannot be verified from the filing, label it unconfirmed. After you finish the table, stop and ask me to approve the output before you proceed to any action that would count as final, such as submitting, sending, deleting, or paying.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.