Planning with Files Tutorial: Persistent Markdown Workflow Memory for AI Coding Agents

June 22, 2026 ยท View on GitHub

Learn how to use OthmanAdi/planning-with-files to run Manus-style file-based planning workflows across Claude Code and other AI coding environments.

GitHub Repo License Release

Why This Track Matters

planning-with-files is one of the fastest-growing workflow skills for coding agents and provides a concrete system for durable planning beyond volatile context windows.

This track focuses on:

  • setting up the skill quickly in Claude Code
  • applying the 3-file planning pattern consistently
  • using hooks, templates, and scripts for reliability
  • adapting the workflow across multiple IDE/agent runtimes

Current Snapshot (auto-updated)

Mental Model

flowchart LR
    A[Task request] --> B[task_plan.md]
    B --> C[findings.md]
    C --> D[progress.md]
    D --> E[Hook-driven reminders]
    E --> F[Completion checks and recovery]

Chapter Guide

ChapterKey QuestionOutcome
01 - Getting StartedHow do I install and launch the workflow quickly?Working baseline
02 - Core Philosophy and the 3-File PatternWhy does filesystem memory improve agent reliability?Strong mental model
03 - Installation Paths Across IDEs and AgentsHow do setup paths vary across toolchains?Better portability
04 - Commands, Hooks, and Workflow OrchestrationHow do commands and hooks enforce discipline?Higher consistency
05 - Templates, Scripts, and Session RecoveryHow do I recover state and keep momentum after context resets?Better resilience
06 - Multi-IDE Adaptation (Codex, Gemini, OpenCode, Cursor)How do I reuse the same planning system across platforms?Cross-platform fluency
07 - Troubleshooting, Anti-Patterns, and Safety ChecksHow do I avoid failure loops and workflow drift?Reliability playbook
08 - Contribution Workflow and Team AdoptionHow do teams standardize and evolve this pattern?Adoption readiness

What You Will Learn

  • how to make agent workflows durable with persistent markdown memory
  • how to run and recover complex tasks with less context loss
  • how to port the planning system across major AI coding IDEs
  • how to govern and extend the workflow in team settings

Source References


Start with Chapter 1: Getting Started.

Full Chapter Map

  1. Chapter 1: Getting Started
  2. Chapter 2: Core Philosophy and the 3-File Pattern
  3. Chapter 3: Installation Paths Across IDEs and Agents
  4. Chapter 4: Commands, Hooks, and Workflow Orchestration
  5. Chapter 5: Templates, Scripts, and Session Recovery
  6. Chapter 6: Multi-IDE Adaptation (Codex, Gemini, OpenCode, Cursor)
  7. Chapter 7: Troubleshooting, Anti-Patterns, and Safety Checks
  8. Chapter 8: Contribution Workflow and Team Adoption

Generated by AI Codebase Knowledge Builder