Automated web trend intelligence parser and breaking AI tools summarizer. Powered by the official Ego Lite (ego-browser skill by citrolabs) rate-limit resilient browser automation and local Ollama LLM synthesis.
Ego Lite AI Trend Monitor tracks trending AI models and tools configured dynamically in config.json.
When web search services or trend pages enforce strict HTTP 429 quota rate limits on automated scraping, this tool uses the official Ego Lite browser task space and rate-limit backoff engine to fetch trend signals, and leverages local Ollama model (gemma4:e2b) to synthesize a clean Markdown report (output.md).
| Component | Technology | Description |
|---|---|---|
| Browser Core | Official ego-browser (citrolabs/ego-lite) |
Resilient agentic browser automation & snapshot engine |
| AI LLM Engine | Ollama (gemma4:e2b) |
Local inference with high speed and zero API quota costs |
| Config & Data Core | Python 3.12 / Standard Lib | Zero heavy dependencies, pure standard library execution |
| Automation | Windows Batch Script (run_automation.bat) |
One-click workflow execution |
.
├── config.json # Target keywords & model parameters
├── trend_fetcher.py # Ego Lite browser trend intelligence collector
├── ai_summarizer.py # Ollama integration engine for AI summary synthesis
├── main.py # Pipeline orchestrator (fetch -> summarize -> output.md)
├── run_automation.bat # Automated 1-click batch script for Windows execution
└── README.md # Project documentation and setup guide
- Daily AI Tools Radar: Automatically scan breaking repositories and model announcements every morning.
- Rate-Limit Circumvention: Continuously fetch search and trend data even under strict quota environments.
- Executive Summary Generation: Distill raw RSS web feeds into structured, actionable Markdown summaries.
- Offline AI Intelligence: Generate zero-cost summaries using local Ollama models without external API keys.
- Competitor Keyword Monitoring: Track brand sentiment and search volume trends for emerging open-source projects.
- Multi-Model Benchmark Integration: Automatically compare performance metrics across newly detected models.
- Automated Discord / Slack Webhooks: Push summary reports directly to team communication channels.
- Sentiment Analysis Layer: Compute community reaction sentiment scores for each tracked keyword.
- Historical Trend Archiving: Persist daily Markdown reports into a searchable SQLite time-series index.
- Custom Prompt Templates: Allow users to define custom Markdown report schemas in
config.json.
Ensure you have Python 3.10+, Ollama, and the official ego-browser skill installed:
# Verify Python installation
python --version
# Verify Ollama model
ollama run gemma4:e2b
# Install official Ego Lite browser skill
npx skills add citrolabs/ego-liteTarget keywords and model parameters are managed exclusively inside config.json:
{
"ollama_model": "gemma4:e2b",
"ollama_url": "http://localhost:11434/api/generate",
"output_file": "output.md",
"keywords": [
"target_keyword_1",
"target_keyword_2"
]
}python main.py.\run_automation.batUpon execution, the system fetches news/trend signals for all configured keywords via ego-browser, passes the gathered context to Ollama (gemma4:e2b), and writes a fresh report at output.md.
To test individual components end-to-end:
# Test trend data collector
python trend_fetcher.py
# Test full pipeline end-to-end
python main.pyDistributed under the MIT License. See LICENSE for more information.
ego lite ego browser ai browser local ai browser playwright alternative citrolabs browser automation web scraper rate limit resilience ollama local llm gemma4:e2b python trend scraper ai tools radar agentic browser