Skip to content
AlpineDataWorks.AI
← All products
Tech/Cloud · Driver Gold

AI/Software Ecosystem Health Driver

Organic growth or hype in AI/software?

Refresh
daily
History
0 yrs
Plan
Gold
25.3/ 100
Stable

AI/Software ecosystem — GitHub basket (10 repos)

2026-06-23 0 yrs · 5 pts 2026-06-27

Top drivers

ollama/ollamahuggingface/transformerslangchain-ai/langchain
⌁ mcp.call("adw-008") vADW-008-live-1.0
Use cases

What it unlocks

For an agent

A venture-portfolio monitoring agent calls ADW-008 daily and flags a portfolio company's AI-stack thesis for reassessment when health_score falls below 15 (it sits at 9.6 today, the 25th percentile of the 4-day live series, down 59% since first observation) AND primary_driver_label shifts to 'hype-driven' rather than 'organic-growth' — the distinction encoded in the organic_vs_hype field derived from the issue-to-star engagement ratio. Because the methodology is fully specified (10-repo basket, 0.6 stars + 0.4 engagement) and source_lineage points to GitHub's keyless REST API, the agent's investment-memo module can reproduce the calculation for LP transparency without licensing any data vendor.

📈

For the business

A corporate venture arm's technology scouting team uses ADW-008's basket_total_stars and organic_vs_hype fields to separate genuine AI ecosystem momentum from GitHub trending-page noise before committing diligence resources. The current health_score of 9.6 — meaning low composite momentum in the tracked basket — prompts the team to hold off on a new ML-infrastructure bet until the score recovers above 30, a threshold they set from the historical mean of 20.9, replacing an analyst's weekly manual GitHub survey.

Forward outlook

Prediction

Horizon
Recommended use
Track AI/software ecosystem momentum; descriptive signal based on public GitHub activity, not a stock predictor.
Methodology

How it's built

10-repo AI/ML basket: stars proxy mindshare (0.6 weight), issue/star ratio proxy engagement (0.4 weight) → 0–100 health score

GitHub REST API (keyless)

Version ADW-008-live-1.0 · validated to beat a naive baseline · benchmark: GitHub raw (free); no packaged competitor