Aller au contenu
TakoAPI
Retour aux skills

finviz-crawler

par camopel

DevOps & Cloud
398Téléchargements rapportés par clawskills.sh
0Vues
0 mention J'aimeConnecte-toi pour aimer ce skill
—Pas encore noté
Ta note

Résumé

Continuous financial news crawler for finviz.com with SQLite storage, article extraction, and query tool.

Description

Ce que fait ce skill

A background daemon that crawls finviz.com every 5 minutes, storing headlines and full article content in a local SQLite database. Includes a query tool to filter by time window, ticker, or content type, with output formatted for LLM summarization pipelines.

Unlike one-shot finance APIs, it accumulates a local historical archive so you can query any past window without rate limits or subscription fees.

Quand l'utiliser

  • Tracking NVDA or TSLA headlines without refreshing a browser
  • Generating a daily 6 AM market briefing via cron
  • Searching three weeks of financial news without an internet connection
  • Building a local archive before paywalled articles disappear
  • Feeding yesterday's headlines into an LLM for a portfolio summary

Exemple de workflow

ENTRÉE

User asks: Summarize NVDA news from the last 24 hours

AGENT
  1. 1

    Runs finviz_query.py with --hours 24 filtered to the NVDA ticker

  2. 2

    Retrieves matching article titles and content from local SQLite

  3. 3

    Formats output as clean text for LLM input

  4. 4

    Passes formatted articles to the LLM for summarization

  5. 5

    Returns a concise summary of NVDA headlines and key themes

SORTIE

A paragraph summarizing NVDA-related news from the past 24 hours, drawn from Bloomberg, Reuters, and Yahoo Finance articles

Installation

clawhub install camopel/finviz-crawler

Exécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.

Liens