Dividend Growth Pullback Screener combines a fundamental dividend screen with a technical filter. A Python script checks yield, market cap, three-year dividend growth, payout sustainability and financial health through the Financial Modeling Prep API, then calculates a 14-period RSI from daily closes and keeps stocks at or below 40.
An optional two-stage mode uses a FINVIZ Elite pre-screen to shrink the candidate list before the detailed FMP analysis. Results are ranked by a composite score and saved as JSON and Markdown. The output is a screening aid and not investment advice.
What You Get
- A ranked list of stocks that meet the dividend growth, yield and RSI criteria
- Per stock: current yield, three-year dividend growth, payout and free cash flow coverage, RSI value and zone
- A composite score weighting dividend growth, financial quality, technical setup and valuation
- A JSON results file and a Markdown screening report
Common Use Cases
- Looking for dividend growers during a market pullback
- Narrowing a large universe with a FINVIZ pre-screen before spending FMP calls
- Relaxing the RSI or growth thresholds when no stock qualifies
- Pairing with market breadth and news skills to confirm a broad pullback
What Makes It Unique
It joins two filters that screeners usually keep apart: strong dividend compounding and short-term oversold timing. The sustainability test is strict: free cash flow must exceed dividends paid for non-REITs, missing cash flow data never confirms sustainability, and REITs are judged on a simplified funds-from-operations payout. The report groups entries by RSI zone (extreme oversold, strong oversold, early pullback).
Under The Hood
The agent runs a Python script that calls the Financial Modeling Prep API over the network with the user’s key, and when requested calls the FINVIZ Elite API with a second key. Both keys come from environment variables or command line options. It uses a bundled client that masks keys in output, retries once on malformed JSON, and waits and retries once on a rate limit. The script writes JSON and Markdown files; depending on the build they land in the working directory or a logs folder, so the agent should check the printed paths. The bundled FMP reference guide also shows an optional shell command that appends the API key to a shell startup file such as .bashrc; the agent only does this if the user asks for a persistent key. Free FMP plans may not include the screener endpoint, and the skill says so.