writing-plans

What This Skill Does

The writing-plans skill is an essential productivity agent designed to bridge the gap between abstract design discussions and concrete implementation. It enforces a rigorous, step-by-step methodology for executing software development tasks, ensuring that every architectural change is broken down into micro-tasks taking 2-5 minutes each. By requiring explicit file paths, implementation snippets, and testable verification commands for every step, this skill eliminates ambiguity in the development lifecycle.

Installation

To integrate this skill into your environment, run the following command within your terminal or OpenClaw interface: clawhub install openclaw/skills/skills/nefas11/writing-plans-2

Use Cases

This skill is perfect for complex refactoring projects, feature development in large codebases, or onboarding new developers to a specific architectural pattern. It is best used immediately after a design brainstorming session has concluded, providing a clear roadmap for the 'build' phase. It prevents 'scope creep' and 'monster tasks' that often lead to bugs or burnout by enforcing granular, verifiable progress.

Example Prompts

1. "We finished the design for the authentication module. Please generate an implementation plan following the writing-plans format."
2. "I need to integrate the Stripe API. Use the writing-plans skill to break this into 3-minute chunks with tests for each step."
3. "Create an implementation plan for adding dark mode to the CSS variables, ensuring all files are clearly specified."

Tips & Limitations

• Granularity is Key: If the AI suggests a task that feels like it will take longer than 5 minutes, force it to split the task further. Smaller chunks lead to higher success rates.
• Verification First: Always ensure the verification command is runnable in your local environment. If the skill generates a test that relies on a tool not currently installed, treat that as a blocker.
• Constraint adherence: The skill works best when you strictly hold it to the 'Anti-Patterns' defined in the documentation. Do not allow vague descriptions; require the AI to write actual code snippets for the implementation phase to ensure clarity.