Codebase 5-Year Forecast Skill

What this skill does

Given a codebase, a GitHub URL, or a project pitch, this skill produces a reproducible, evidence-backed 5-year forecast of the project's defensibility. The output is a clinical, quantified verdict — not a flattering essay — that survives being re-run by different models or tools and arrives at the same numerical conclusion.

The core principle: the LLM gathers data and writes the narrative; a deterministic Python script does the scoring. Same inputs → same scores → same verdict, every time.

When to invoke

Trigger this skill when the user wants any of:

• A long-horizon (multi-year) outlook on a codebase, app, product, or technical idea
• A moat / defensibility analysis with quantified scoring
• An AI-disruption risk assessment for a software product
• A pivot recommendation backed by data
• A "will my project still be relevant in N years" question
• A "should I keep building this" strategic review

Do not invoke for: short-term roadmap planning, bug triage, code review, single-feature cost estimates, or general architecture advice. Those are different tasks.