Claw-Screener: skill pack for Hermes Agent

Stock screener skill combining Williams %R oversold scans with Buffett-style analysis for US and Thai markets

What Claw-Screener does

Claw-Screener is a stock screener that combines technical and fundamental analysis. It looks for oversold stocks with the Williams %R indicator and scores quality with Warren Buffett-style formulas, using SEC data, for the US S&P 500 and the Thai SET market. Four scripts cover combined screening, a technical-only scan, single-stock analysis and a Compounding Machine screen that applies Carlson filters such as ROIC above 15 percent and operating margin above 20 percent.

The repository includes an Agent Skills-compatible SKILL.md. For Hermes Agent, clone the full repository into ~/.hermes/skills/finance/claw-screener, since the skill needs the TypeScript sources, or add an existing clone through skills.external_dirs in config.yaml. Then use /claw-screener in chat or ask Hermes to screen or analyze stocks. Results can be printed as text, JSON or a Telegram-formatted message, and SEC lookups are cached in a SQLite file.

Key features

Williams %R oversold screening with an adjustable threshold
Buffett-style scoring from 0 to 10 using ten formulas
Single-stock analysis such as AAPL or PTT.BK
Compounding Machine screen with ROIC, margin and buyback filters
Text, JSON and Telegram output formats
US S&P 500 and Thai SET market universes

When to use it

Asking Hermes to list oversold quality stocks in the S&P 500
Running a quick fundamental check on one ticker
Sending a screening result to a Telegram chat in the Telegram format

Who it is for: Retail investors and developers who want a stock-screening skill for Hermes Agent or OpenClaw.

How it fits with Hermes Agent

The README has a Hermes Agent section that installs the skill under ~/.hermes/skills/finance/ and invokes it with /claw-screener. It also supports OpenClaw, and the repository is tagged hermes-skill.

How to install Claw-Screener

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

git clone https://github.com/rsoutar/claw-screener.git ~/.hermes/skills/finance/claw-screener
cd ~/.hermes/skills/finance/claw-screener && npm install

Requirements: A full clone of the repository (the skill needs its TypeScript sources) and npm install; US fundamental analysis uses SEC data

Note: Thai market screening is technical-only because it does not use SEC data.

FAQ

What is Claw-Screener?

Claw-Screener is a stock screener for US and Thai markets. It combines Williams %R oversold signals with Warren Buffett-style fundamental scoring and ships a SKILL.md so AI agents can run it.

Does Claw-Screener work with Hermes Agent?

Yes. Clone the repository into ~/.hermes/skills/finance/claw-screener and run npm install, then use /claw-screener in chat. You can also point skills.external_dirs in ~/.hermes/config.yaml at an existing clone.

Is Claw-Screener free and open source?

Yes. The repository is released under the MIT license.