How does the 3-Strike Error Protocol improve task completion?

The protocol prevents the AI from repeating failed actions. It forces a diagnostic fix on attempt one, an alternative method on attempt two, and a total rethink on attempt three, ensuring efficient problem-solving.

Why use file-based planning instead of just the chat context?

While the chat context is limited and volatile (like RAM), file-based planning acts as a persistent disk. This ensures that goals, findings, and progress remain accessible even during long tasks or context refreshes.

When is the best time to use the planning-with-files skill?

It is recommended for any task requiring more than five tool calls, research-heavy projects, building new features, or any multi-step workflow where organization is critical.

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

The 2-Action Rule requires the AI to save key findings to a text file after every two view, browser, or search operations. This prevents visual or multimodal information from being lost if the context window shifts.

Manus-Style Persistent Planning

Name: Manus-Style Persistent Planning
Author: OthmanAdi

byOthmanAdi

•

11,059

생산성 및 워크플로

Implements a robust, file-based workflow for Claude Code to handle complex tasks using persistent markdown files as external memory.

소개

This skill equips Claude Code with a professional-grade planning framework inspired by the Manus workflow pattern. By maintaining persistent files—task_plan.md, findings.md, and progress.md—Claude overcomes the limitations of volatile context windows and ensures critical data is never lost during long-running operations. It is designed specifically for high-complexity tasks, deep research, and multi-step engineering projects where maintaining a clear 'working memory on disk' is essential for success and error recovery.

주요 기능

Standardized templates for task plans, research findings, and session logs
11,059 GitHub stars
Decision matrices to optimize when to read vs. write to the filesystem
The 2-Action Rule for immediate logging of multimodal findings
Integrated 3-Strike Error Protocol for systematic troubleshooting
Persistent 'working memory' that survives context window resets

사용 사례

Conducting deep research tasks requiring numerous browser or tool calls
Managing complex, multi-step software engineering projects
Debugging intricate system errors that require structured attempt tracking

소개

주요 기능

Standardized templates for task plans, research findings, and session logs
11,059 GitHub stars
Decision matrices to optimize when to read vs. write to the filesystem
The 2-Action Rule for immediate logging of multimodal findings
Integrated 3-Strike Error Protocol for systematic troubleshooting
Persistent 'working memory' that survives context window resets

사용 사례

Conducting deep research tasks requiring numerous browser or tool calls
Managing complex, multi-step software engineering projects
Debugging intricate system errors that require structured attempt tracking