Skip to content
Agentmart
Submit a tool

plan

Break down a feature, bugfix, or task into a concrete implementation plan before touching code.

Skill by Kotivskyi

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/kotivskyi-plan

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Break down a feature, bugfix, or task into a concrete implementation plan before touching code. Use at the start of any non-trivial piece of work to clarify scope, surface dependencies, and produce an ordered checklist the user can approve before implementation begins.

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/plan

What's inside

Skill name
plan
Description
Break down a feature, bugfix, or task into a concrete implementation plan before touching code. Use at the start of any non-trivial piece of work to clarify scope, surface dependencies, and produce an ordered checklist the user can approve before implementation begins.
License
MIT
Instructions
55 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Plan

Produce a structured implementation plan for the work the user described.
Stop before writing any code — the deliverable is the plan itself.

## Steps

### 1. Understand the goal

Restate the objective in one sentence. If the request is ambiguous, ask the
minimum questions needed to sharpen it before proceeding.

### 2. Explore the affected area

Read only what is needed to understand scope:

- entry points, relevant types, and interfaces in the area being changed
- existing tests covering related behaviour
- any config, migration, or schema files that would be touched

Do not read the whole codebase. Stop when you have enough to write the plan.

### 3. Identify constraints and risks

Note anything that could cause the plan to fail or expand scope unexpectedly:

- external dependencies (APIs, services, packages)
- backwards-compatibility requirements
- performance or security implications
- anything that must stay unchanged

### 4. Write the plan

Structure it as:

**Goal** — one sentence.

**Scope** — what is in and what is out.

**Steps** — ordered list. Each step should be small enough to verify
independently. Mark steps that must happen in sequence with a dependency note.

**Open questions** — anything the user must decide before or during
implementation (keep this list short; only genuine forks belong here).

**Verification** — how to confirm the work is correct when done (tests to
run, manual checks, etc.).

### 5. Get approval

Present the plan and wait. Do not start implementing until the user explicitly
approves (or adjusts and approves). If they approve, hand off to the
`pr-watch` skill once a PR exists, or run the implementation inline.

From Kotivskyi/skills, MIT.

More from Kotivskyi/skills

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 16, 2026 (1 month ago)
Last release
Jun 22, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Kotivskyi

Cite this page

plan skill. Agent Mart. https://agentmart.dev/skills/kotivskyi-plan (last updated 2026-10-24)