plan

Writes actionable markdown implementation plans with bite-sized tasks to .dag/plans/ without executing code.

1|Updated Jun 17, 2026
One-click install
npx skills add https://github.com/specdog/collar --skill plan-specdog
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: plan
Source: https://github.com/specdog/collar/tree/main/skills/plan
Command: npx skills add https://github.com/specdog/collar --skill plan-specdog

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Jumping straight into implementation without a verified, structured plan leads to wrong assumptions, vague tasks, and wasted effort. This Skill produces a concrete, step-by-step implementation plan saved as a markdown file, so execution becomes obvious and delegable. ## Core Features & Use Cases - Plan-only mode: Inspects the repository with read-only commands and writes a plan without editing project files, running mutating commands, or committing. - Structured plan format: Generates plans with goal, architecture, exact file paths, complete copy-pasteable code, exact test commands with expected output, and bite-sized 2-5 minute tasks following TDD and frequent commits. - Verification-first planning: Requires every factual claim (duplication, compilation failures, missing tests) to be verified with actual tool output before being written into the plan. - Use Case: Before implementing a multi-step feature, ask for a plan; the Skill explores the codebase, designs the approach, and saves a timestamped plan under .dag/plans/ ready for task-by-task execution via subagent-driven-development. ## Quick Start Ask the agent to plan the implementation of your feature, for example: plan how to add password reset to the auth module and save it under .dag/plans/.

Frequently Asked Questions about plan

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

FAQPage Schema
How do I create an implementation plan before writing code?▼

Invoke plan mode and describe the feature; the agent inspects the repository with read-only commands, designs the approach, and saves a markdown plan under .dag/plans/ with a timestamped filename. The plan contains bite-sized tasks with exact file paths, complete code, and test commands.

What should a good implementation plan include?▼

A good plan includes a goal, architecture summary, tech stack, and sequential tasks of 2-5 minutes each. Every task lists exact files to create or modify, complete copy-pasteable code, exact test commands with expected output, and a commit step.

Does plan mode modify or execute any project files?▼

No. Plan mode only writes the plan markdown file under .dag/plans/ and uses read-only commands to inspect the repository. It does not edit project files, run mutating terminal commands, commit, push, or perform external actions.

Why do AI-generated plans fail on unfamiliar codebases?▼

Plans fail when built on unverified claims, such as assuming files are duplicates or that code does not compile. Every factual claim must be verified with actual tool output like diff, grep, or build commands before being written into the plan.

How do I execute a saved plan after planning?▼

Use the subagent-driven-development skill to execute the plan task-by-task. Each task is dispatched to a fresh subagent with full context, followed by a spec compliance review and a code quality review before proceeding.