How do I install the Implementation Planner skill?

Run npx skillmds@latest add flowerf19/implementation-planner in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Implementation Planner safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Implementation Planner?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Implementation Planner free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Implementation Planner?

Flowerf19 (@flowerf19) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Implementation Planner by flowerf19

Name: Implementation Planner
Author: Flowerf19

File contents

Create or update an execution-ready plan another agent can follow. Planning-only: write the plan artifact, not runtime code or config.

Before drafting

If .agents/ is missing, run architecture-docs first to bootstrap it.
Read project instructions, the request, any existing plan, and relevant repo evidence.
Resolve decisions that affect scope, interfaces, sequencing, or acceptance criteria.
Ask only when no safe default exists; otherwise record the assumption.
If the spec conflicts with the repo, name the conflict and propose a compatible alternative. Stop for unsafe or impossible requirements.

Plan content

Include:

Goal and measurable success criteria.
Phase goals (GOAL-NNN) with small independently verifiable tasks (TASK-NNN).
Interfaces, schemas, commands, and files only where needed to remove execution ambiguity.
Correctness-relevant edge cases and failure modes.
Focused tests and acceptance criteria.
Assumptions and defaults.

Do not include broad rewrites, speculative abstractions, TODO/TBD placeholders, file inventories, or call-graph dumps.

Format

Default path: .agents/plans/<slug>.md.

---
status: draft
created: YYYY-MM-DD
last_updated: YYYY-MM-DD
---

Body sections: Summary, Tasks, Test Plan, Assumptions.

### GOAL-001: <phase goal>

| ID | Task | Done | Date |
|----|------|------|------|
| TASK-001 | <small verifiable step> | | |

IDs are append-only; never renumber existing TASK or GOAL entries.

Lifecycle

draft - waiting for approval.
in-progress - approved and being executed.
done - implementation merged; keep as a record or move to .agents/decisions/ for a lasting decision.
abandoned - superseded or cancelled; keep a one-line reason.

Update existing plans in place. Keep completed IDs, append new IDs, strike superseded tasks with a reason, bump last_updated, and restore in-progress when approved work resumes.

---name: implementation-plannerdescription: Create or update execution-ready implementation plans from specs, features, bugs, or subsystem goals, grounded in repository evidence.---Create or update an execution-ready plan another agent can follow. Planning-only: write the plan artifact, not runtime code or config.## Before drafting- If `.agents/` is missing, run `architecture-docs` first to bootstrap it.- Read project instructions, the request, any existing plan, and relevant repo evidence.- Resolve decisions that affect scope, interfaces, sequencing, or acceptance criteria.- Ask only when no safe default exists; otherwise record the assumption.- If the spec conflicts with the repo, name the conflict and propose a compatible alternative. Stop for unsafe or impossible requirements.## Plan contentInclude:- Goal and measurable success criteria.- Phase goals (`GOAL-NNN`) with small independently verifiable tasks (`TASK-NNN`).- Interfaces, schemas, commands, and files only where needed to remove execution ambiguity.- Correctness-relevant edge cases and failure modes.- Focused tests and acceptance criteria.- Assumptions and defaults.Do not include broad rewrites, speculative abstractions, TODO/TBD placeholders, file inventories, or call-graph dumps.## FormatDefault path: `.agents/plans/<slug>.md`.```yaml---status: draftcreated: YYYY-MM-DDlast_updated: YYYY-MM-DD---```Body sections: Summary, Tasks, Test Plan, Assumptions.```md### GOAL-001: <phase goal>| ID | Task | Done | Date ||----|------|------|------|| TASK-001 | <small verifiable step> | | |```IDs are append-only; never renumber existing `TASK` or `GOAL` entries.## Lifecycle- `draft` - waiting for approval.- `in-progress` - approved and being executed.- `done` - implementation merged; keep as a record or move to `.agents/decisions/` for a lasting decision.- `abandoned` - superseded or cancelled; keep a one-line reason.Update existing plans in place. Keep completed IDs, append new IDs, strike superseded tasks with a reason, bump `last_updated`, and restore `in-progress` when approved work resumes.

Frequently asked questions

Implementation Planner