Yapay zeka becerisiMonitor CompetitorsMarketing

Competitor Intel — daily competitor signal feed — Claude Skill

Claude Code için bir Claude becerisi · Gooseworks — çalıştır: /competitor-intel (Claude'da)·Güncellendi: 10 Nis 2026

Şununla uyumluClaude·ChatGPT·OpenClaw

Automated competitor monitoring across Reddit, X, and LinkedIn

  • Builds deep competitor profiles via one-time agent research
  • Runs daily Reddit and Twitter signal collection
  • Runs weekly LinkedIn deep-dive with founder activity
  • Generates daily and weekly digest reports
  • Delivers reports as markdown files and via email

Kim için

Ne yapar

Always-on competitor monitoring

Replace ad-hoc Google searches with a structured daily and weekly competitive intel feed.

Catch buying-signal moments

Surface competitor frustrations on Reddit and Twitter that you can convert to outbound.

Track founder narratives

Watch what competitor founders are saying publicly to understand strategic direction.

Nasıl çalışır

1

Take competitor list as input via setup script

2

Run agent-driven deep research to populate profiles

3

Run daily Reddit and Twitter scrapers for ongoing signals

4

Run weekly LinkedIn deep dive on Mondays

5

Generate and email daily and weekly digest reports

İyileştirdiği metrikler

Lead Generation
More qualified leads from converting competitor frustration signals into outbound
Marketing

Uyumlu araçlar

Competitor Intelligence kullanmak ister misiniz?

Nasıl başlamak istediğinizi seçin.

Claude Code'da çalıştır
Ücretsiz. Açık kaynak.

Bu beceriyi bilgisayarınıza yerel olarak kurun ve çalıştırın.

1
Claude Code'u kur

Bilgisayarınızda bir terminal açın ve şu komutu yapıştırın:

2
Beceriyi kur

Bu, beceriyi tüm dosyalarıyla bilgisayarınıza indirir:

Tüm projelerinizde kullanılabilir hale getirmek için sona -g ekleyin.

3
Çalıştırın

Claude Code'u başlatın, ardından komutu yazın:

sonra
Kaynağı GitHub'da görüntüle
ElasticFlow'da kullan
Ekip ve işbirliği özellikleri

Becerileri tarayıcınızdan çalıştırın. Sonuçları paylaşın, erişimi yönetin, ekibinizle işbirliği yapın. Terminal gerekmez.

14 gün ücretsiz deneme. Dilediğiniz zaman iptal edin.

Competitor Intelligence

Automated competitor monitoring and intelligence gathering system.

Important: Before running any competitor intel commands, ask the user which competitors they want to track. Do not assume or hardcode competitor names.

Quick Start

One-Time: Research a New Competitor

# Create a new competitor profile (agent-driven research)
python3 competitor-intel/scripts/setup_competitor.py --name "CompanyName" --website https://example.com --slug companyname

Then run an agent research session to populate the profile.

Daily: Reddit + Twitter Monitoring

# Run daily signals collection (automated via cron)
python3 competitor-intel/scripts/run_daily.py

# Or for a specific competitor
python3 competitor-intel/scripts/run_daily.py --competitor <competitor-slug>

Weekly: Deep Dive + LinkedIn

# Run weekly deep dive (Monday mornings)
python3 competitor-intel/scripts/run_weekly.py

Generate Report

# Generate daily or weekly report
python3 competitor-intel/scripts/generate_report.py --type daily --date 2026-02-21
python3 competitor-intel/scripts/generate_report.py --type weekly --date 2026-02-21

Data Structure

  • competitor-intel/competitors/[slug]/profile.md - Deep research profile per competitor
  • competitor-intel/competitors/[slug]/snapshots/ - Point-in-time weekly snapshots
  • competitor-intel/reports/ - Daily and weekly digest reports
  • competitor-intel/raw-data/ - Raw JSON from Reddit/Twitter scrapes
  • competitor-intel/config.json - Global configuration

Tracked Competitors

Competitors are configured per-project. Ask the user which competitors to track, then add them here:

CompetitorSlugStatus
(ask user)

Automation

  • Daily (8 AM PT): Reddit + Twitter scrape → daily report → email
  • Weekly (Monday 9 AM PT): Deep dive + LinkedIn + weekly report → email

Dependencies

  • Apify API (Reddit + Twitter scraping) - token in config.json
  • Web search (LinkedIn founder tracking, one-time research)
  • AgentMail (report delivery)