Does this skill work after using the /clear command?

Yes. The skill includes a session-catchup script specifically designed to recover context and sync the AI's state with existing files after a session reset.

What is Manus-style planning?

It is a workflow pattern that treats the filesystem as external 'RAM' for the AI. By writing plans and findings to disk, the AI avoids the limitations of its volatile context window during long tasks.

Where are the planning files stored?

All planning files (task_plan.md, findings.md, and progress.md) are created and maintained directly in your project's root directory, not in the skill folder.

When should I use the Planning with Files skill?

It is recommended for any task requiring more than five tool calls, research projects, or complex feature implementations where keeping goals in the 'attention window' is critical.

What is the 3-Strike Error Protocol?

It is a systematic approach to debugging where the AI must diagnose the fix on attempt 1, try an alternative method on attempt 2, and perform a total rethink on attempt 3 before asking the user for help.

Planning with Files

Name: Planning with Files
Author: OthmanAdi

byOthmanAdi

•

11,059

Productivity & Workflow

Implements persistent, Manus-style markdown planning to manage complex coding tasks and systematic research through file-based state.

About

This skill equips Claude Code with a robust 'working memory' by utilizing the local filesystem to store task phases, research findings, and session progress. By maintaining persistent files like task_plan.md and findings.md, it prevents context loss during long-running operations, ensures project goals remain within the attention window, and enforces a rigorous 3-strike error protocol. It is specifically designed for high-complexity projects requiring more than five tool calls, featuring automatic session recovery and systematic debugging through structured, project-level documentation.

Key Features

Pre-tool and post-tool hooks to ensure planning files stay synchronized
Standardized 3-strike error protocol to prevent repetitive AI failure loops
11,059 GitHub stars
Automatic session recovery and context catchup after /clear or restarts
Persistent state management using task_plan.md, findings.md, and progress.md
Template-based initialization for research, development, and debugging workflows

Use Cases

Managing multi-step software development projects that span dozens of tool calls
Executing complex debugging sessions that require logging multiple failed attempts and resolutions
Conducting deep-dive research tasks requiring persistent storage of web-fetched data

About

Key Features

Pre-tool and post-tool hooks to ensure planning files stay synchronized
Standardized 3-strike error protocol to prevent repetitive AI failure loops
11,059 GitHub stars
Automatic session recovery and context catchup after /clear or restarts
Persistent state management using task_plan.md, findings.md, and progress.md
Template-based initialization for research, development, and debugging workflows

Use Cases

Managing multi-step software development projects that span dozens of tool calls
Executing complex debugging sessions that require logging multiple failed attempts and resolutions
Conducting deep-dive research tasks requiring persistent storage of web-fetched data