software-effort-estimation

To estimate software development effort, the Skill analyzes git history and cloc line counts to calculate reproducible person-month estimates. It combines codebase sizing with multi-model approaches like COCOMO II to generate a defensible cost and timeline range.

The best way to value a codebase is by combining infrastructure inventory parsing, test code measurement, and multi-model estimation. This produces a realistic effort baseline that quantifies traditional engineering costs versus AI-assisted delivery patterns.

Yes, you can compare traditional engineering effort against AI-assisted development. The Skill compares estimated baseline effort with actual git delivery patterns to quantify productivity gains and calculate speedup across the analyzed history.

Yes, you need cloc, git, and bc installed to calculate person-months. These dependencies are required to perform cloc-based line counting, parse git history, and execute multi-model calculations for the final estimation report.

Software effort estimation works for infrastructure-heavy systems by applying infrastructure premiums and parsing infrastructure inventory. It adjusts multi-model calculations to accurately reflect the effort baseline of complex infrastructure-heavy codebases.

A limitation of using COCOMO II for codebase valuation is false precision. To mitigate this, the Skill applies five estimation approaches and encourages cross-checking with repeated counts, spot checks, and model convergence for a reliable range.