Claude News Fetcher - Media Monitoring System

Workspace template for conducting media monitoring and gathering articles into a well-defined structured data format.

Purpose

This repository serves two primary use cases:

1. Media Monitoring: Track coverage about specific subjects across various publications
2. Research Data Collection: Gather and organize source materials for research projects

Quick Start

After Cloning This Repository

IMPORTANT: If you're using this repository for media monitoring, run the setup command first:

/setup-monitoring

This interactive command will:

• Gather details about what you're monitoring (person, company, topic, etc.)
• Understand your monitoring objectives and scope
• Identify key topics, publications, and relevance criteria
• Update the repository's CLAUDE.md file with this context

Running this setup ensures Claude understands your monitoring goals and can better assist with article fetching, classification, and analysis throughout your project.

Basic Usage

Once setup is complete, you can start fetching articles:

Fetch a single article:

/fetch [URL]

Fetch multiple articles:

/batch [URL1] [URL2] [URL3]

Analyze your collection:

/analyze

Features

Article Organization

• Date-based structure: Articles organized by year and month (/articles/YYYY/MM_MonthName/)
• Consistent naming: article_theme_DDMMYY_publication.md format
• Rich metadata: Publication info, dates, authors, keywords, summaries
• JSON schema validation: Ensures data quality and consistency

Available Commands

Command	Description
/setup-monitoring	Configure monitoring context (run this first!)
/fetch	Fetch and save a single article
/batch	Process multiple articles at once
/analyze	Generate statistics and insights about your collection
/search	Search articles by keyword, publication, or date
/stats	Show collection statistics
/validate	Validate articles against schema
/export	Export articles in various formats (CSV, JSON)
/profile-pub	Research and profile a publication

Specialized Agents

Located in .claude/agents/:

• article-fetcher: Fetch and save individual articles with metadata
• batch-fetch: Process multiple articles efficiently
• article-analyzer: Analyze trends and generate reports
• publication-profiler: Research publication metadata and bias
• archive-organizer: Maintain archive quality and consistency

Directory Structure

/articles/              # All fetched articles (organized by date)
  /YYYY/               # Year folders
    /MM_MonthName/     # Month folders
      article_*.md     # Individual article markdown files
/schemas/              # Data schemas and validation rules
  /fetched-article/
    fetch-schema.json       # JSON schema for article data
    fetch-instructions.md   # Detailed fetching guidelines
/.claude/              # Claude Code configuration
  /agents/             # Specialized subagents
  /commands/           # Slash commands
/exports/              # Exported data (CSV, JSON, etc.)
/reports/              # Generated analysis reports
/publication-profiles/ # Publication metadata and profiles

Article Data Schema

Each article includes:

Required fields:

• Title, publication name, URL
• Full text content and word count
• Claude-generated summary
• Classification (coverage/op-ed/by-subject)
• Keywords (5 tags)
• Display summary

Optional enrichment:

• Author information (name, title, bio)
• Publication details (summary, editorial leaning)
• Comment analysis (if article has comments)

See /schemas/fetched-article/fetch-schema.json for complete schema.

Workflow Examples

Single Article Fetch

User: "Fetch this article: https://economist.com/example"
Claude: [Fetches, extracts, validates, saves to correct date folder]

Batch Processing

User: "Fetch these 10 articles about climate policy"
Claude: [Processes all articles, provides summary report]

Analysis

User: "/analyze the last month"
Claude: [Generates statistics, trends, topic breakdown]

Search

User: "/search climate policy"
Claude: [Returns matching articles with metadata]

Quality Standards

• Content Quality: Clean extraction, proper markdown formatting
• Metadata Quality: Accurate dates, authors, classifications
• Organizational Quality: Consistent naming and structure
• Schema Compliance: All articles validated against JSON schema

Error Handling

The system gracefully handles:

• Paywall-protected articles
• Invalid URLs
• Missing metadata (uses sensible defaults)
• Publication date ambiguity
• Schema validation issues

Integration

Works with Daniel's ecosystem:

• MCP Servers: Context7, Resend, Time
• Notion: Export summaries and reports
• Obsidian: Reference in notes
• Analytics: Export for visualization

Best Practices

1. Run /setup-monitoring first - Establishes monitoring context
2. Use batch operations - More efficient for multiple articles
3. Validate regularly - Run /validate to check data quality
4. Analyze periodically - Generate insights with /analyze
5. Profile publications - Build metadata with /profile-pub
6. Export for backup - Use /export for data preservation

Technical Details

• Language: Markdown for articles, JSON for schemas
• Validation: JSON Schema compliance checking
• Organization: Date-based hierarchical structure
• Automation: Specialized agents for complex workflows
• Extensibility: Add custom agents and commands as needed

Getting Help

• Check CLAUDE.md for comprehensive system documentation
• Review /schemas/fetched-article/fetch-instructions.md for fetching guidelines
• Use /stats to understand your current collection
• Ask Claude for help with any command or workflow

Version Control

This repository is designed to be version controlled. Commit logical groups of articles with clear messages:

git add articles/2025/11_Nov/
git commit -m "Add 15 climate policy articles from Nov 2025"
git push

---

Ready to start? Run /setup-monitoring to configure your media monitoring workspace.

---

To view an index of my Claude Code related projects, click here