When is it best to initialize a new planning session?

You should use this skill whenever you are starting a complex task, a research project, or any operation expected to take more than five tool calls.

What is the 2-Action Rule included in this skill?

The 2-Action Rule requires saving key findings to text files after every two browser or search operations, preventing important data from being buried in the chat history.

Is this skill compatible with different operating systems?

Yes, it includes specialized scripts for both Unix-like systems (macOS/Linux) via Bash and Windows environments via PowerShell.

How does the session recovery feature work?

The skill includes a catchup script that analyzes git diffs and current planning files to restore the AI's state after a session restart or workspace clear.

Why use file-based planning instead of standard chat?

Files provide 'Disk' memory that persists across /clear commands and prevents context window exhaustion, ensuring the AI never loses track of long-term goals.

Manus-Style File Planning

Name: Manus-Style File Planning
Author: OthmanAdi

byOthmanAdi

•

13,549

•

생산성 및 워크플로

Implements a persistent markdown-based planning system to manage complex multi-step tasks and maintain long-term AI memory.

This skill enhances Claude's capability to handle complex, long-running tasks by treating the local filesystem as a persistent 'disk' memory. By maintaining structured files like task_plan.md, findings.md, and progress.md, it prevents context loss during long sessions and ensures consistent execution for deep research or large-scale development projects. It includes automated session recovery to sync context after restarts and enforces disciplined workflow patterns like the '2-action rule' to ensure no discovery is lost, making it an essential tool for any task requiring significant tool usage or multi-phase planning.

주요 기능

01Structured templates for phase tracking, research findings, and session logs

02Cross-platform initialization scripts for macOS, Linux, and Windows

03Persistent markdown-based memory system for high-complexity tasks

04Disciplined 2-action rule for frequent and reliable data persistence

05Automatic session recovery and context syncing after /clear commands

0613,549 GitHub stars

사용 사례

01Tracking multi-step debugging processes to ensure errors are logged and never repeated

02Managing complex refactoring projects that span multiple AI sessions

03Conducting deep research and documentation tasks requiring high context retention

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add othmanadi/planning-with-files planning-with-files

For use in Claude.ai and ChatGPT

Download Skill