planning-with-files
Planning with Files
This skill implements a persistent memory workflow using three markdown files. This allows you to handle complex, multi-step tasks without losing context or forgetting goals.
The 3-File Pattern
For any complex task, you must maintain these three files in the root of the project (or a specific plans/ directory if preferred by the user):
task_plan.md: The master plan. Tracks phases, specific tasks, and their completion status.findings.md: The knowledge base. Stores lessons learned, decisions made, and technical details discovered.progress.md: The session log. A chronological record of actions taken, results observed, and thoughts.
When to Use
- When the user asks for "planning with files".
- When starting a complex feature or refactoring.
- When the task requires multiple tool calls and steps.
- When you need to "remember" things for future sessions.
Instructions
1. Initialization
At the start of a task, check if these files exist. If not, create them by copying from the templates in .claude/templates/ or using the content below.
- task_plan.md: Break down the user's request into high-level phases and granular tasks. Mark them as
[ ](pending). - findings.md: Initialize with an empty structure for "Key Decisions" and "Technical Learnings".
- progress.md: Add a header for the current session (Date/Time).
2. Execution Loop
Before and after running tools, update the files:
- Update
progress.md: specific actions you are about to take or have just taken. Record command outputs or test results briefly. - Update
findings.md: If you discover how a system works, a hidden bug, or make an architectural decision, record it here immediately. - Update
task_plan.md: When a sub-task is done, mark it[x]. If new tasks are discovered, add them.
3. Context Management
- Do not rely on your context window for long-term memory.
- Read
task_plan.mdto know "what's next". - Read
findings.mdto know "how things work" or "what we decided". - Read
progress.mdto know "what did I just do".
Templates
task_plan.md Template
# Task Plan: [Task Name]
## Status
- [ ] Phase 1: [Name]
- [ ] Task 1.1
- [ ] Task 1.2
- [ ] Phase 2: [Name]
## Current Context
- Focus: [Current Focus]
- Blockers: [Any Blockers]
findings.md Template
# Findings & Knowledge
## Key Decisions
- [Date] [Decision]: [Rationale]
## Technical Learnings
- [Tool/Library]: [How to use it / Gotchas]
- [Architecture]: [Description]
## Unresolved Questions
- [ ] [Question]?
progress.md Template
# Progress Log
## [YYYY-MM-DD HH:MM] Session Start
- Goal: [Goal for this session]
### [HH:MM] Action
- executed: `command`
- result: success/fail
- thoughts: ...
More from chudiren/ai-agent-testing-platform
ui-ux-pro-max
UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 9 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient. Integrations: shadcn/ui MCP for component search and examples.
14code-review-router
Intelligently routes code reviews between Gemini CLI and Codex CLI based on tech stack, complexity, and change characteristics. Use when you want an automated code review of your current changes.
14skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
12dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
9finishing-a-development-branch
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
9verification-before-completion
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
9