Why use files instead of just chatting with Claude?

Chat context is volatile and can be lost or 'forgotten' during long sessions. Writing to files like task_plan.md ensures critical discoveries and project goals are preserved even if the session is cleared.

When should I use the Planning with Files skill?

Use it for any task requiring more than five tool calls, complex research, or projects involving multiple files where keeping track of progress is critical.

Does this skill work with existing projects?

Yes, it is designed to create planning files directly in your active project directory, allowing for seamless integration with your existing codebase and Git history.

How does session recovery work?

The skill includes specialized scripts that scan your directory for existing planning files and Git diffs to help Claude quickly re-orient itself after a restart or memory reset.

What is a Manus-style workflow?

It is a methodology where the AI uses persistent files on the local disk as 'working memory' to manage complex tasks that exceed the standard context window limitations.

Manus-Style Task Planning

Name: Manus-Style Task Planning
Author: iseepatterns-jz

byiseepatterns-jz

Productivity & Workflow

Implements persistent, file-based planning to manage complex multi-step tasks and maintain context during long AI coding sessions.

About

Planning with Files brings a highly effective Manus-style workflow to Claude Code by using markdown files as persistent 'working memory on disk.' By maintaining dedicated files for task plans, research findings, and session progress, this skill overcomes the limitations of volatile context windows, ensuring the AI never loses sight of the ultimate goal. It features built-in session recovery, structured error-tracking protocols, and a rigorous 'Read vs Write' decision matrix to keep your engineering and research projects organized and on track.

Key Features

Standardized 3-strike error protocol to prevent repetitive logic failures.
0 GitHub stars
Automatic session recovery and context catchup after clearing memory.
The 2-Action Rule for immediate saving of multimodal research data.
Real-time phase tracking and milestone logging for complex workflows.
Persistent state management using task_plan.md, findings.md, and progress.md.

Use Cases

Executing multi-step software engineering tasks that require more than 5 tool calls.
Starting new projects from scratch where architectural goals must remain in focus.
Performing deep-dive research projects where visual or browser data must be preserved.

About

Key Features

Standardized 3-strike error protocol to prevent repetitive logic failures.
0 GitHub stars
Automatic session recovery and context catchup after clearing memory.
The 2-Action Rule for immediate saving of multimodal research data.
Real-time phase tracking and milestone logging for complex workflows.
Persistent state management using task_plan.md, findings.md, and progress.md.

Use Cases

Executing multi-step software engineering tasks that require more than 5 tool calls.
Starting new projects from scratch where architectural goals must remain in focus.
Performing deep-dive research projects where visual or browser data must be preserved.