Slack Poster Skill

This skill posts business news and reports to Slack channels with intelligent topic analysis and threaded organization for better readability.

Purpose

• Post business news digests with automatic topic grouping
• Provide big-picture insights by correlating related stories
• Use threaded messages to organize content by topic
• Send alerts, notifications, or reports to internal Slack channels programmatically

Key Features

1. Topic Analysis & Grouping

• Automatically analyzes headlines to identify major themes
• Groups related stories together (e.g., "AI & Technology", "Markets & Economy")
• Correlates between stories to show the big picture
• Counts stories per topic for quick overview

2. Threaded Message Format

• Main message: Summary showing topic breakdown and key insights
• Thread replies: Detailed stories organized by topic
• Each topic thread includes all relevant stories with source attribution

3. Topic Categories

The skill automatically classifies stories into:

• 🤖 AI & Technology
• 📈 Markets & Economy
• 👔 Corporate Leadership & Governance
• 🛒 Retail & Consumer
• 💼 Business & Finance
• 🏥 Healthcare & Biotech
• 💼 Workplace & Careers
• 📰 Other News

Implementation

Basic Usage (post_to_slack.py)

Simple format posting top headlines with blocks formatting.

Advanced Usage (post_to_slack_threaded.py)

Intelligent topic analysis with threaded organization:

• Analyzes all headlines for common themes
• Groups stories by topic
• Creates main summary message with big-picture insights
• Posts detailed stories in topic-specific threads

Authentication

Slack credentials:

• Token: Use environment variable SLACK_TOKEN
• Channel: stock_ops_dev (or use environment variable SLACK_CHANNEL)

Example Output

Main Message:

📰 Business News Digest - November 19, 2025

40 stories analyzed across 7 major topics

🔍 Today's Big Picture:
• AI & Technology: 19 stories
• Markets & Economy: 4 stories
• Corporate Leadership & Governance: 2 stories

🤖 AI dominates headlines with 19 stories covering Nvidia earnings,
   OpenAI board changes, and the AI bubble debate

Thread Replies: Each topic gets its own detailed thread with all related stories.

Output Structure

All outputs are organized in the outputs/ directory:

• Reports should go into outputs/<agent_name>/<customer_name>/reports/
• Scripts should go into outputs/<agent_name>/<customer_name>/scripts/
• Raw outputs (csvs, jsons) should go into outputs/<agent_name>/<customer_name>/raw/
• Screenshots should go into outputs/<agent_name>/<customer_name>/screenshots/