Skip to content

feO2x/guided-coding / claude-plugin/claude-skills/write-plan

write-plan

Unverified

Write a Guided Coding plan or follow-up plan after its approach has been discussed. Run only when explicitly requested by the user or by another Guided Coding skill.

Write a Plan

Your goal is to write an Initial Plan or Follow-Up Plan into the ai-plans/ folder at the repository root. After writing, report the plan path and stop. When another Guided Coding skill invoked this one, return to it instead of stopping.

Plan File Name

Name the draft <short-title>.md, using a shortened kebab-case version of the ticket title. If no ticket title is available, derive one from the discussion. Ask the user for a title only if you cannot derive one.

Plan Structure

Use the first plan structure that applies:

  1. A plan structure the user names when invoking this skill: a path to a Markdown file, or the name of a file in assets/ relative to this skill file, such as rdat-plan-structure.md.
  2. A plan structure that ai-plans/AGENTS.md names in the same way or describes itself.
  3. assets/default-plan-structure.md relative to this skill file.

A bare filename refers to a file in assets/ relative to this skill file. Resolve other relative paths from the repository root. If the selected file does not exist, stop and report it. Do not fall back to another structure.

What Happens After Writing a Plan?

The user reviews the draft and may request changes. Once you and the user agree it is ready for implementation, it can be frozen. The only permitted edits after freezing are checking off acceptance criteria.

Freezing renames the file to <timestamp>-<ticket-id>-<short-title>.md, or <timestamp>-<short-title>.md when there is no ticket ID. It also adds a *Frozen at ...* line directly below the title. Do not freeze the plan while running this skill.

Follow-up Plans

A Follow-Up Plan supersedes decisions of a Frozen Plan. The decision to write one is made in the Guiding Phase, when the review of the implementation calls for larger changes.

Use the plan structure selected above and the same file naming convention. Read every earlier document for the ticket. In the opening section of the plan (## Rationale in the default structure), name the plans this one follows by exact filename.

Record only the changed decisions and outcomes. State exactly which earlier decisions this plan supersedes and why. Do not repeat unchanged decisions or contradict earlier plans silently.

Most Importantly

Keep plans concise. Use plain language.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 467kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 319kUpdated Claude CodeCodexVerifiedInstall