topic-monitor

Topic Monitor

Monitor what matters. Get notified when it happens.

Topic Monitor transforms your assistant from reactive to proactive by continuously monitoring topics you care about and intelligently alerting you only when something truly matters.

Core Capabilities

Topic Configuration - Define subjects with custom parameters
Scheduled Monitoring - Automated searches at configurable intervals
AI Importance Scoring - Smart filtering: immediate alert vs digest vs ignore
Contextual Summaries - Not just links—meaningful summaries with context
Weekly Digest - Low-priority findings compiled into readable reports
Memory Integration - References your past conversations and interests

First Run

When you first use Topic Monitor, run the interactive setup wizard:

python3 scripts/setup.py

The wizard will guide you through:

Topics - What subjects do you want to monitor?
Search queries - How to search for each topic
Keywords - What terms indicate relevance
Frequency - How often to check (hourly/daily/weekly)
Importance threshold - When to send alerts (low/medium/high)
Weekly digest - Compile non-urgent findings into a summary

The wizard creates config.json with your preferences. You can always edit it later or use manage_topics.py to add/remove topics.

Example session:

🔍 Topic Monitor - Setup Wizard

What topics do you want to monitor?
  > AI Model Releases
  > Security Vulnerabilities
  > 

--- Topic 1/2: AI Model Releases ---
  Search query for 'AI Model Releases' [AI Model Releases news updates]: new AI model release announcement

Quick Start

Already know what you're doing? Here's the manual approach:

Topic Configuration

Each topic has:

name - Display name (e.g., "AI Model Releases")
query - Search query (e.g., "new AI model release announcement")
keywords - Relevance filters (["GPT", "Claude", "Llama", "release"])
frequency - hourly, daily, weekly
importance_threshold - high (alert immediately), medium (alert if important), low (digest only)
channels - Where to send alerts (["telegram", "discord"])
context - Why you care (for AI contextual summaries)

Example config.json

Scripts

manage_topics.py

Manage research topics:

monitor.py

Main monitoring script (run via cron):

How it works:

Reads topics due for checking (based on frequency)
Searches using web-search-plus or built-in web_search
Scores each result with AI importance scorer
High-importance → immediate alert
Medium-importance → saved for digest
Low-importance → ignored
Updates state to prevent duplicate alerts

digest.py

Generate weekly digest:

Output format:

setup_cron.py

Configure automated monitoring:

Creates cron entries:

AI Importance Scoring

The scorer uses multiple signals to decide alert priority:

Scoring Signals

HIGH priority (immediate alert):

Major breaking news (detected via freshness + keyword density)
Price changes >10% (for finance topics)
Product releases matching your exact keywords
Security vulnerabilities in tools you use
Direct answers to specific questions you asked

MEDIUM priority (digest-worthy):

Related news but not urgent
Minor updates to tracked products
Interesting developments in your topics
Tutorial/guide releases
Community discussions with high engagement

LOW priority (ignore):

Duplicate news (already alerted)
Tangentially related content
Low-quality sources
Outdated information
Spam/promotional content

Learning Mode

When enabled (learning_enabled: true), the system:

Tracks which alerts you interact with
Adjusts scoring weights based on your behavior
Suggests topic refinements
Auto-adjusts importance thresholds

Learning data stored in .learning_data.json (privacy-safe, never shared).

Memory Integration

Topic Monitor connects to your conversation history:

Example alert:

🔔 Dirac Live Update

Version 3.8 released with the room correction improvements you asked about last week.

Context: You mentioned struggling with bass response in your studio. This update includes new low-frequency optimization.

[Link] | [Full details]

How it works:

Reads references/memory_hints.md (create this file)
Scans recent conversation logs (if available)
Matches findings to past context
Generates personalized summaries

memory_hints.md (optional)

Help the AI connect dots:

Alert Channels

Requires OpenClaw message tool:

Discord

Webhook-based:

Email

SMTP or API:

Advanced Features

Alert Conditions

Fine-tune when to alert:

Regex Patterns

Match specific patterns:

Rate Limiting

Prevent alert fatigue:

State Management

.research_state.json

Tracks:

Last check time per topic
Alerted URLs (deduplication)
Importance scores history
Learning data (if enabled)

Example:

.findings/ directory

Stores digest-worthy findings:

Best Practices

Start conservative - Set importance_threshold: medium initially, adjust based on alert quality
Use context field - Helps AI generate better summaries
Refine keywords - Add negative keywords to filter noise: "keywords": ["AI", "-clickbait", "-spam"]
Enable learning - Improves over time based on your behavior
Review digest weekly - Don't ignore the digest—it surfaces patterns
Combine with personal-analytics - Get topic recommendations based on your chat patterns

Integration with Other Skills

web-search-plus

Automatically uses intelligent routing:

Product/price topics → Serper
Research topics → Tavily
Company/startup discovery → Exa

personal-analytics

Suggests topics based on conversation patterns:

"You've asked about Rust 12 times this month. Want me to monitor 'Rust language updates'?"

Privacy & Security

All data local - No external services except search APIs
State files gitignored - Safe to use in version-controlled workspace
Memory hints optional - You control what context is shared
Learning data stays local - Never sent to APIs

Troubleshooting

No alerts being sent:

Check cron is running: crontab -l
Verify channel config (Telegram chat ID, Discord webhook)
Run with --dry-run --verbose to see scoring

Too many alerts:

Increase importance_threshold
Add rate limiting
Refine keywords (add negative filters)
Enable learning mode

Missing important news:

Decrease importance_threshold
Increase check frequency
Broaden keywords
Check .research_state.json for deduplication issues

Digest not generating:

Verify .findings/ directory exists and has content
Check digest cron schedule
Run manually: python3 scripts/digest.py --preview

Example Workflows

Track Product Release

Monitor Competitor

Research Topic

Credits

Built for ClawHub. Uses web-search-plus skill for intelligent search routing.

Topic Monitor

Monitor what matters. Get notified when it happens.

Topic Monitor transforms your assistant from reactive to proactive by continuously monitoring topics you care about and intelligently alerting you only when something truly matters.

Core Capabilities

Topic Configuration - Define subjects with custom parameters
Scheduled Monitoring - Automated searches at configurable intervals
AI Importance Scoring - Smart filtering: immediate alert vs digest vs ignore
Contextual Summaries - Not just links—meaningful summaries with context
Weekly Digest - Low-priority findings compiled into readable reports
Memory Integration - References your past conversations and interests

First Run

When you first use Topic Monitor, run the interactive setup wizard:

python3 scripts/setup.py

The wizard will guide you through:

Topics - What subjects do you want to monitor?
Search queries - How to search for each topic
Keywords - What terms indicate relevance
Frequency - How often to check (hourly/daily/weekly)
Importance threshold - When to send alerts (low/medium/high)
Weekly digest - Compile non-urgent findings into a summary

The wizard creates config.json with your preferences. You can always edit it later or use manage_topics.py to add/remove topics.

Example session:

🔍 Topic Monitor - Setup Wizard

What topics do you want to monitor?
  > AI Model Releases
  > Security Vulnerabilities
  > 

--- Topic 1/2: AI Model Releases ---
  Search query for 'AI Model Releases' [AI Model Releases news updates]: new AI model release announcement

Quick Start

Already know what you're doing? Here's the manual approach:

Topic Configuration

Each topic has:

name - Display name (e.g., "AI Model Releases")
query - Search query (e.g., "new AI model release announcement")
keywords - Relevance filters (["GPT", "Claude", "Llama", "release"])
frequency - hourly, daily, weekly
importance_threshold - high (alert immediately), medium (alert if important), low (digest only)
channels - Where to send alerts (["telegram", "discord"])
context - Why you care (for AI contextual summaries)

Example config.json

Scripts

manage_topics.py

Manage research topics:

monitor.py

Main monitoring script (run via cron):

How it works:

Reads topics due for checking (based on frequency)
Searches using web-search-plus or built-in web_search
Scores each result with AI importance scorer
High-importance → immediate alert
Medium-importance → saved for digest
Low-importance → ignored
Updates state to prevent duplicate alerts

digest.py

Generate weekly digest:

Output format:

setup_cron.py

Configure automated monitoring:

Creates cron entries:

AI Importance Scoring

The scorer uses multiple signals to decide alert priority:

Scoring Signals

HIGH priority (immediate alert):

Major breaking news (detected via freshness + keyword density)
Price changes >10% (for finance topics)
Product releases matching your exact keywords
Security vulnerabilities in tools you use
Direct answers to specific questions you asked

MEDIUM priority (digest-worthy):

Related news but not urgent
Minor updates to tracked products
Interesting developments in your topics
Tutorial/guide releases
Community discussions with high engagement

LOW priority (ignore):

Duplicate news (already alerted)
Tangentially related content
Low-quality sources
Outdated information
Spam/promotional content

Learning Mode

When enabled (learning_enabled: true), the system:

Tracks which alerts you interact with
Adjusts scoring weights based on your behavior
Suggests topic refinements
Auto-adjusts importance thresholds

Learning data stored in .learning_data.json (privacy-safe, never shared).

Memory Integration

Topic Monitor connects to your conversation history:

Example alert:

🔔 Dirac Live Update

Version 3.8 released with the room correction improvements you asked about last week.

Context: You mentioned struggling with bass response in your studio. This update includes new low-frequency optimization.

[Link] | [Full details]

How it works:

Reads references/memory_hints.md (create this file)
Scans recent conversation logs (if available)
Matches findings to past context
Generates personalized summaries

memory_hints.md (optional)

Help the AI connect dots:

Alert Channels

Requires OpenClaw message tool:

Discord

Webhook-based:

Email

SMTP or API:

Advanced Features

Alert Conditions

Fine-tune when to alert:

Regex Patterns

Match specific patterns:

Rate Limiting

Prevent alert fatigue:

State Management

.research_state.json

Tracks:

Last check time per topic
Alerted URLs (deduplication)
Importance scores history
Learning data (if enabled)

Example:

.findings/ directory

Stores digest-worthy findings:

Best Practices

Start conservative - Set importance_threshold: medium initially, adjust based on alert quality
Use context field - Helps AI generate better summaries
Refine keywords - Add negative keywords to filter noise: "keywords": ["AI", "-clickbait", "-spam"]
Enable learning - Improves over time based on your behavior
Review digest weekly - Don't ignore the digest—it surfaces patterns
Combine with personal-analytics - Get topic recommendations based on your chat patterns

Integration with Other Skills

web-search-plus

Automatically uses intelligent routing:

Product/price topics → Serper
Research topics → Tavily
Company/startup discovery → Exa

personal-analytics

Suggests topics based on conversation patterns:

"You've asked about Rust 12 times this month. Want me to monitor 'Rust language updates'?"

Privacy & Security

All data local - No external services except search APIs
State files gitignored - Safe to use in version-controlled workspace
Memory hints optional - You control what context is shared
Learning data stays local - Never sent to APIs

Troubleshooting

No alerts being sent:

Check cron is running: crontab -l
Verify channel config (Telegram chat ID, Discord webhook)
Run with --dry-run --verbose to see scoring

Too many alerts:

Increase importance_threshold
Add rate limiting
Refine keywords (add negative filters)
Enable learning mode

Missing important news:

Decrease importance_threshold
Increase check frequency
Broaden keywords
Check .research_state.json for deduplication issues

Digest not generating:

Verify .findings/ directory exists and has content
Check digest cron schedule
Run manually: python3 scripts/digest.py --preview

Example Workflows

Track Product Release

Monitor Competitor

Research Topic

Credits

Built for ClawHub. Uses web-search-plus skill for intelligent search routing.

Documentation

Documentation

Topic Monitor

Core Capabilities

First Run

Quick Start

Topic Configuration

Example config.json

Scripts

manage_topics.py

monitor.py

digest.py

setup_cron.py

AI Importance Scoring

Scoring Signals

Learning Mode

Memory Integration

memory_hints.md (optional)

Alert Channels

Telegram

Discord

Email

Advanced Features

Alert Conditions

Regex Patterns

Rate Limiting

State Management

.research_state.json

.findings/ directory

Best Practices

Integration with Other Skills

web-search-plus

personal-analytics

Privacy & Security

Troubleshooting

Example Workflows

Track Product Release

Monitor Competitor

Research Topic

Credits

Documentation

topic-monitor

Documentation

Topic Monitor

Core Capabilities

First Run

Quick Start

Topic Configuration

Example config.json

Scripts

manage_topics.py

monitor.py

digest.py

setup_cron.py

AI Importance Scoring

Scoring Signals

Learning Mode

Memory Integration

memory_hints.md (optional)

Alert Channels

Telegram

Discord

Email

Advanced Features

Alert Conditions

Regex Patterns

Rate Limiting

State Management

.research_state.json

.findings/ directory

Best Practices

Integration with Other Skills

web-search-plus

personal-analytics

Privacy & Security

Troubleshooting

Example Workflows

Track Product Release

Monitor Competitor