planning-with-files

Track multi-step AI agent tasks using persistent markdown plan files.

Updated May 9, 2026
One-click install
npx skills add https://github.com/kk20300113-png/my-claude-skills --skill planning-with-files-kk20300113-png
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: planning-with-files
Source: https://github.com/kk20300113-png/my-claude-skills/tree/main/planning-with-files
Command: npx skills add https://github.com/kk20300113-png/my-claude-skills --skill planning-with-files-kk20300113-png

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill eliminates the risk of losing track of goals, progress, and critical discoveries during long, multi-step AI tasks that require dozens of tool calls, where volatile context windows cause important information to be forgotten between interactions.

Core Features & Use Cases

  • Persistent Phase Tracking: Creates and maintains a task_plan.md file to break complex work into manageable phases, track completion status, and log errors to avoid repeating failed actions.
  • Research & Discovery Storage: Uses findings.md to capture research results, technical decisions, and multimodal content (like browser results or image findings) immediately after discovery, before context is lost.
  • Session Recovery & Parallel Workflows: Automatically recovers unsynced context after a /clear command, and supports isolated plan directories for working on multiple independent tasks in the same repository simultaneously.
  • Use Case: Ideal for tasks like technical research, feature development, complex bug fixes, or any project requiring 5 or more tool calls across multiple sessions.

Quick Start

Ask the AI to use the planning-with-files skill to break down your multi-step project into tracked phases, capture all research findings and errors, and maintain a session log for seamless recovery after context resets.

Frequently Asked Questions about planning-with-files

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

FAQPage Schema
How do I prevent context loss during multi-step AI agent workflows?

To prevent context loss during multi-step AI agent workflows, use persistent markdown files like task_plan.md and findings.md as external working memory to track phases, log errors, and store research discoveries.

How do I recover my AI agent session progress after a context reset?

You can recover AI agent session progress after a context reset by using a skill that automatically reads persistent markdown planning files to resync lost context, restoring your exact phase tracking and previous discoveries.

What is the best way to manage parallel task isolation in AI agent workflows?

The best way to manage parallel task isolation in AI agent workflows is by creating isolated plan directories within your repository, allowing multiple independent multi-step tasks to maintain separate progress tracking.

Does file-based planning work for complex bug fixes across multiple sessions?

Yes, file-based planning works effectively for complex bug fixes across multiple sessions by capturing technical decisions and multimodal research in markdown files before volatile context windows lose the information.

When should I use persistent markdown files for AI task planning?

You should use persistent markdown files for AI task planning when your project requires 5 or more tool calls across multiple sessions, such as technical research, feature development, or complex bug fixes.