skills.rest
Clay-HHK
Clay-HHKCOMMUNITY
Explore all skills by this author →
GitHub

planning-with-files

Creates persistent markdown files for task planning, progress tracking, and knowledge storage.

1|Updated Mar 5, 2026
Search Tags:#planning#markdown#context engineering#agent workflow#progress tracking#notes#task tracking
One-click install
npx skills add https://github.com/Clay-HHK/claude-skills --skill planning-with-files-clay-hhk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: planning-with-files
Source: https://github.com/Clay-HHK/claude-skills/tree/main/planning-with-files
Command: npx skills add https://github.com/Clay-HHK/claude-skills --skill planning-with-files-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Complex multi-step tasks cause AI agents to lose track of goals after many tool calls, forget errors, and stuff context with large outputs. This Skill solves that by using persistent markdown files as external working memory. ## Core Features & Use Cases - 3-File Pattern: Creates task_plan.md for phases and progress, notes.md for research findings, and a deliverable file for final output. - Attention Refresh: Re-reads the plan file before each decision to keep goals in the attention window across long sessions. - Error Logging: Records every failure in an Errors Encountered section to build knowledge and avoid silent retries. - Use Case: When asked to research a topic and write a report, the agent first writes task_plan.md with phases, stores findings in notes.md during research, then synthesizes the final deliverable while updating progress after each phase. ## Quick Start Ask the agent to plan this complex task using markdown files to track phases, notes, and progress before starting work.

Frequently Asked Questions about planning-with-files

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

FAQPage Schema
How do I keep an AI agent on track during long multi-step tasks?▼

Create a task_plan.md file with phases and checkboxes, then re-read it before each major decision. This keeps goals in the attention window even after dozens of tool calls, preventing the agent from forgetting its original objective.

What is the 3-file pattern for agent task planning?▼

The 3-file pattern uses task_plan.md to track phases and status, notes.md to store research findings, and a deliverable file for final output. The plan is updated after each phase and read before each decision.

When should I use file-based planning versus simple answers?▼

Use the 3-file pattern for tasks with three or more steps, research tasks, or anything spanning multiple tool calls. Skip it for simple questions, single-file edits, and quick lookups where planning overhead adds no value.

Why should agents log errors instead of retrying silently?▼

Logging errors in the plan file preserves failure traces that help the model update its understanding and avoid repeated mistakes. Silent retries waste tokens and lose learning, while visible error records support genuine recovery behavior.

What are the limitations of markdown file planning for agents?▼

File-based planning adds overhead for trivial tasks and depends on the agent consistently reading and updating the files. It also requires filesystem access, so it does not apply in environments without file read and write tools.

Related Skills

Explore More Software Engineering Skills→
khideskhides

ralph-plan

Generate a structured JSON state file from user goals for Ralph automation loops.

Community
Advanced
pelcherspelchers

research-docs-session

Coordinate ADR phase planning, validation, and archival for research sessions.

Community
Intermediate
EremesNGEremesNG
1

verification

Run automated checks and manual review to verify completed work.

Community
Intermediate
ealvar3zealvar3z

taskwarrior-task-management

Manage Taskwarrior tasks scoped to the current Git project.

Community
Intermediate