harness-progress

Track session progress and state using JSON logs and Git commits.

Updated Apr 12, 2026
One-click install
npx skills add https://github.com/Taek-D/kbo_appintoss --skill harness-progress
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-progress
Source: https://github.com/Taek-D/kbo_appintoss/tree/main/miniapp/.claude/skills/harness-progress
Command: npx skills add https://github.com/Taek-D/kbo_appintoss --skill harness-progress

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Harness Progress Skill provides a disciplined, session-scoped management tool for AI-driven mini-app development, enforcing one focus per session and tracking state across sessions with claude-progress.txt and feature_list.json.

Core Features & Use Cases

  • Per-session focus: ensures only one feature is worked on at a time to maintain quality.
  • Cross-session state tracking: logs progress and decisions via claude-progress.txt and feature_list.json for seamless continuation and auditing.
  • Git-based rollback: commits completed features to enable safe revert if issues arise.

Quick Start

Start a new session by reading claude-progress.txt and feature_list.json to determine the next feature, then begin work on the first incomplete item.

Frequently Asked Questions about harness-progress

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

FAQPage Schema
How do I track AI development progress across multiple sessions?

To track AI development progress across multiple sessions, use a harness pattern that logs state via JSON-based progress logging and feature lists. This approach maintains per-session focus and records incremental feature completion for seamless continuation and auditing.

What is the best way to manage feature implementation state for AI mini-apps?

Managing feature implementation state for AI mini-apps requires enforcing one focus per session and logging decisions via JSON-based progress tracking. This ensures deterministic feature completion and provides clear handoff between development sessions.

How do I safely rollback changes from an AI-driven development session?

Safely rollback changes from an AI-driven development session by committing completed features to Git. This Git-based rollback mechanism enables safe reverts if issues arise during incremental mini-app feature development.

Can I resume an incomplete feature across different AI development sessions?

Yes, you can resume an incomplete feature across different AI development sessions by reading progress tracking logs and feature lists at the start. These files determine the next incomplete feature to work on.

How does a harness pattern enforce single-task focus for AI coding agents?

A harness pattern enforces single-task focus for AI coding agents by restricting each session to one incomplete feature from the feature list. This prevents context dilution and maintains quality during incremental mini-app development.

Do I need Git to audit progress in incremental mini-app development?

Yes, Git is needed to audit progress in incremental mini-app development. Git commits provide auditable state changes and enable deterministic rollback, while JSON-based progress logging tracks decisions for cross-session continuity.