pi-planning-with-files

Implements file-based planning with task_plan.md, findings.md, and progress.md for multi-step tasks.

1|Updated Apr 3, 2026
One-click install
npx skills add https://github.com/Pgooone/oh-pgone-claudecode --skill pi-planning-with-files-pgooone
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pi-planning-with-files
Source: https://github.com/Pgooone/oh-pgone-claudecode/tree/main/.claude/plugins/cache/planning-with-files/planning-with-files/2.23.0/.pi/skills/planning-with-files
Command: npx skills add https://github.com/Pgooone/oh-pgone-claudecode --skill pi-planning-with-files-pgooone

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? AI agents lose track of goals during long tasks because context windows are volatile and limited. This Skill persists plans, findings, and progress to markdown files on disk, preventing goal drift, lost research, and repeated errors across sessions. ## Core Features & Use Cases - Three-File Planning Pattern: Creates task_plan.md for phase tracking, findings.md for research storage, and progress.md for session logging in your project directory. - Error and Recovery Protocols: Enforces the 2-Action Rule for saving findings, the 3-Strike Error Protocol, and a 5-Question Reboot Test to verify context integrity. - Session Recovery: The session-catchup.py script detects unsynced context from previous sessions after a context reset, so work can resume without losing state. - Use Case: When asked to research a topic and build a deliverable over many tool calls, the agent writes a plan first, logs discoveries after every search, and recovers state automatically after /clear. ## Quick Start Use the planning-with-files skill to create task_plan.md, findings.md, and progress.md for this project before starting work.

Frequently Asked Questions about pi-planning-with-files

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use file-based planning for complex AI agent tasks?▼

Create three markdown files in your project root before starting: task_plan.md for phases and goals, findings.md for research, and progress.md for session logs. Update them after each phase and re-read the plan before major decisions.

How to recover AI agent context after clearing the session?▼

Run the session-catchup.py script with your project path. It scans the previous session transcript, finds the last planning file update, and reports unsynced messages so you can update your planning files and resume.

Does planning-with-files work with Claude Code hooks?▼

Hooks like PreToolUse, PostToolUse, and Stop are Claude Code specific and are not supported in Pi Agent. The core planning pattern, templates, rules, and session-catchup.py script work in Pi Agent.

When should I not use file-based task planning?▼

Skip it for simple questions, single-file edits, and quick lookups. The pattern is designed for multi-step tasks with three or more steps, research work, or anything spanning many tool calls.

What is the 2-Action Rule in agent planning?▼

After every two view, browser, or search operations, immediately save key findings to a text file. This prevents visual and multimodal information from being lost when the context window resets.