Contributing

Contributions welcome! Here's how to help.

Key Areas

Calibration Data

If you track estimated vs. actual effort using this skill, share anonymized results. This is the most valuable contribution — real data improves the default ratios for everyone.

Format:

| Task Type | Size | PERT Expected | Actual | Delta |
|-----------|------|---------------|--------|-------|
| coding | M | 4 hrs | 4.8 hrs | +20% |
| data-migration | L | 14 hrs | 18 hrs | +29% |

Tracker Mappings

Add support for trackers not yet covered:

• Asana
• Notion
• Shortcut
• Trello
• Azure DevOps

Follow the format in references/output-schema.md — provide both native and embedded field mappings.

Task Types

Propose new task type multipliers for work categories not yet covered. Include:

• The task type name
• Proposed multiplier with rationale
• At least 3 data points from real projects

Formula Improvements

Propose changes to the core formulas. Requirements:

• Research citation or empirical data supporting the change
• Show how the change affects the 5-task benchmark (dark mode, DB migration, Slack service, CSV export, E2E tests)
• Run the regression eval before and after

Evals

Add test cases for:

• Edge cases (very small tasks, very large tasks)
• Specific task types (data-migration, infrastructure)
• Multi-team scenarios
• Different confidence levels
• Different org sizes

How to Submit

1. Fork the repo
2. Create a branch for your change
3. Run the regression eval (evals/eval-regression.md)
4. Submit a PR with:
   • What you changed
   • Why (data, research, or user feedback)
   • Eval results before and after

What Not to Change

• Don't adjust formulas without data or research to support the change
• Don't add complexity for hypothetical scenarios
• Don't break the progressive loading structure (SKILL.md must stay lean)
• Don't add dependencies — the skill is pure markdown by design

Questions?

Open an issue at github.com/Enreign/progressive-estimation/issues.