Research and Screen Stocks With Claude and FMP
Connect Financial Modeling Prep data to Claude and use five reusable prompts to verify market data, compare earnings calls, evaluate fundamentals, form balanced investment theses, and screen
Download the complete Claude and FMP stock research guide
Research and Screen Stocks With Claude and FMP
Connect market data, inspect company fundamentals, compare earnings calls, and build a disciplined undervaluation screen.
What you will achieve
You will configure Claude to use FMP data and run five reusable prompts for researching an individual stock or screening for potentially undervalued companies.
Who this is for
Investors and market researchers who want a faster, structured way to investigate public companies with Claude.
Difficulty: Beginner
Short tutorial
Connect market data, inspect company fundamentals, compare earnings calls, and build a disciplined undervaluation screen.
Guide: Research Stocks in Claude With the FMP Connector
Time: 15 minutes
Get 20% off your FMP subscription:
<https://site.financialmodelingprep.com/pricing-plans?couponCode=miles>
What you will build
This tutorial shows you how to connect FMP to Claude, then use it to retrieve market information, analyze fundamentals, compare earnings call transcripts, construct an investment thesis, and screen for potentially undervalued stocks.
Use the prompts as research templates, replacing the bracketed fields with your company, ticker, sector, and criteria. Market data can be incomplete, delayed, or interpreted incorrectly, so confirm important figures and dates before making a decision. This workflow supports research, not personalized financial advice or guaranteed returns.
1. Connect FMP to Claude
- Open Claude.
- Go to Settings.
- Select Connectors.
- Find and add the connector named FMP.
- Complete any sign in or authorization steps shown.
- Start a new conversation for your stock research.
The source material indicates that the connector starts working with free access, while a paid FMP plan provides more data for heavier use. Exact availability, coverage, and refresh timing are not specified, so check the data returned for each request.
2. Confirm that current market data is available
Before doing deeper analysis, test the connection with one stock. This prompt asks Claude to identify the data date instead of presenting an undated number as current.
Using the connected FMP data, research [COMPANY NAME] ([TICKER]). Report: 1. The latest available share price. 2. The date and time associated with that price, if available. 3. The analyst consensus price target. 4. The number of analysts or estimates behind the consensus, if available. 5. The implied percentage difference between the latest price and the consensus target. Show the calculation for the percentage difference. Clearly label any missing, delayed, or unavailable data. Do not estimate values that are not present in the connected data.
A useful response should distinguish the latest available price from the analyst target. Treat a consensus target as an opinion aggregate, not a prediction.
3. Compare the last two earnings calls
An earnings call comparison can reveal changes in management's confidence, priorities, risks, and expectations. Use this prompt to separate direct statements from Claude's interpretation.
Using the connected data, review the two most recent available earnings call transcripts for [COMPANY NAME] ([TICKER]). Compare management's commentary across the two calls. Focus on: 1. Revenue growth and demand. 2. Margins and costs. 3. Guidance or forward expectations. 4. Capital spending and investment priorities. 5. Product, customer, or geographic trends. 6. Risks mentioned by management. Create a table with these columns: Topic | Earlier call | Latest call | What changed | Supporting statement Then summarize: 1. Where management became more optimistic. 2. Where management became more cautious. 3. Any new issue that appeared in the latest call. 4. Any prior concern that was reduced or removed. Identify the dates of both transcripts. Distinguish direct management statements from your interpretation. If a transcript is unavailable, say so rather than filling the gap.
Pay particular attention to changes in wording. A shift from firm guidance to broad or conditional language can matter even if the headline numbers remain similar.
4. Generate a balanced investment thesis
A useful thesis should explain both why the stock might work and what could invalidate the idea. Ask Claude to connect claims to measurable evidence.
Build a balanced investment thesis for [COMPANY NAME] ([TICKER]) using the latest available FMP data. Structure the analysis as follows: A. Business overview Explain how the company makes money and identify its main business drivers. B. Bull case List the three strongest reasons the business or stock could outperform. Support each reason with available financial or operating data. C. Bear case List the three strongest risks or reasons the stock could underperform. Support each with available evidence. D. Valuation Summarize the available valuation measures and compare them with the company's own historical figures or relevant peers only when that comparison is available. E. Upcoming signals List the next financial or operating indicators that would strengthen or weaken the thesis. F. Thesis breakers Identify specific developments that would invalidate the bull case. For every important number, include its reporting period or data date. Separate facts from interpretation, and clearly identify missing data.
5. Inspect the underlying fundamentals
Run a dedicated fundamentals review rather than relying only on price targets or valuation multiples.
Analyze the underlying fundamentals of [COMPANY NAME] ([TICKER]) using the latest available connected FMP data. Cover these areas: 1. Revenue level and growth trend. 2. Gross, operating, and net margin trends, where available. 3. Earnings and earnings per share trends. 4. Operating cash flow and free cash flow trends. 5. Cash, debt, and overall balance sheet position. 6. Share count changes or dilution, where available. 7. Relevant valuation ratios available in the data. Use a table with these columns: Metric | Latest value | Prior comparable value | Direction | Interpretation After the table, provide: 1. Three fundamental strengths. 2. Three fundamental concerns. 3. The single metric that deserves the closest monitoring. Use comparable reporting periods when discussing changes. Do not treat a missing metric as zero. State the date or period for every key figure.
6. Screen for potentially undervalued stocks
The transcript suggests using fundamentals to find underrated or undervalued companies. Because the source does not supply a fixed screening formula, define your own universe and criteria explicitly. This prevents Claude from quietly changing what “undervalued” means.
Using connected FMP data, screen for potentially undervalued stocks within [MARKET, INDEX, SECTOR, OR COMPANY LIST]. Apply these criteria where the required data is available: 1. Positive revenue or earnings trend over [PERIOD]. 2. Positive operating cash flow or free cash flow. 3. A manageable debt position relative to the company's cash generation. 4. Valuation measures below the relevant sector, peer group, or the company's own historical level, when comparable data is available. 5. No obvious deterioration in the latest reported fundamentals. Return up to [NUMBER] candidates in a table with these columns: Ticker | Company | Latest price and date | Valuation evidence | Fundamental evidence | Main risk | Missing data Rank candidates by the strength of the combined valuation and fundamental evidence, not by price decline alone. Then explain: 1. Why each candidate may be undervalued. 2. What could make the apparent discount misleading. 3. Which metric or upcoming result should be checked next. Do not call a stock undervalued when the available data is insufficient. Do not invent peer comparisons, historical averages, or analyst estimates.
Example screening specification
Replace the bracketed section with a clear scope such as:
Screen companies in the semiconductor sector. Return no more than five candidates. Prefer companies with positive free cash flow and improving revenue or margins. Exclude any company when the available data is too incomplete to assess those conditions.
This is more reliable than asking Claude to find “cheap stocks” without defining quality, valuation, or the research universe.
7. Validate the result before using it
Run this final checklist on every analysis:
- Confirm the ticker and company name.
- Check the date and reporting period attached to each major number.
- Distinguish trailing results from forecasts and analyst targets.
- Verify that comparisons use equivalent periods.
- Look for missing transcripts, estimates, or financial fields.
- Separate management statements from Claude's interpretation.
- Review both the bull case and the bear case.
- Treat a low valuation as a research lead, not proof of undervaluation.
- Confirm critical figures independently before acting.
The strongest workflow is sequential: verify the data connection, inspect the latest market information, compare management commentary, review fundamentals, construct both sides of the thesis, and only then assess whether the valuation appears attractive.
