estimate

Records and refines phase forecasts using workspace state, Git logs, and session-derived durations.

1|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/heymishy/skills-repo --skill estimate-heymishy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: estimate
Source: https://github.com/heymishy/skills-repo/tree/main/.github/skills/estimate
Command: npx skills add https://github.com/heymishy/skills-repo --skill estimate-heymishy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams standardize and accelerate the estimation of the outer loop from discovery through definition, by recording rough forecasts, refining them as scope is known, and calibrating against actuals to improve future planning.

Core Features & Use Cases

  • Record rough estimates at discovery (E1) and refine at definition (E2) with live adjustments (E3).
  • Reconstruct focus time from session logs to produce consistent, comparable metrics across features.
  • Generate normalisation data to improve planning accuracy and surface process improvements.

Quick Start

Trigger /estimate at the end of discovery or /definition to record initial forecasts and begin normalisation.

Frequently Asked Questions about estimate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I track and refine feature estimates across discovery and definition phases?

Feature estimates are tracked by recording rough forecasts at discovery and refining them at definition. This approach reconciles forecasted calendar days with actuals to calibrate and improve future delivery planning.

How does state-tracking and Git log analysis calculate actual development time?

State-tracking and Git log analysis reconstruct focus time by reading workspace state files and JSONL transcripts. This mechanism derives phase durations to produce consistent, comparable metrics across features.

What's the best way to normalize planning data from session logs for future forecasting?

Normalizing planning data is best achieved by reconciling initial forecasts with reconstructed actuals from session logs. This generates normalization data to improve planning accuracy and surface workflow improvements.

Do I need workspace state files and JSONL transcripts to use automated estimate calibration?

Yes, automated estimate calibration requires access to workspace state.json, Git logs, and JSONL transcripts. These dependencies are necessary to derive phase durations and calculate actual focus time.

When should I trigger estimate recording to begin the normalization process?

You should trigger estimate recording at the end of the discovery or definition phases. This captures initial forecasts and begins the normalization process by comparing them against actual delivery times.

Can I adjust phase estimates live as project scope changes during delivery?

Yes, phase estimates can be adjusted live as scope changes during delivery. This live adjustment capability allows teams to refine initial forecasts and maintain accurate planning data throughout the workflow.