Skip to content
Agentmart
Submit a tool

plan

Turn conversation context into a Product Requirements Document (PRD) and break it down into independently-grabbable tracking issues.

Skill by drlove2002

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/drlove2002-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

Turn conversation context into a Product Requirements Document (PRD) and break it down into independently-grabbable tracking issues. Use when the user wants to plan a feature, write a spec, or break a project into tickets.

Use cases: documents, writing

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
Turn conversation context into a Product Requirements Document (PRD) and break it down into independently-grabbable tracking issues. Use when the user wants to plan a feature, write a spec, or break a project into tickets.
License
MIT
Instructions
104 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Plan

Synthesize the conversation and codebase context into a PRD, then break it into vertical-slice tracking issues.

## 1. Context

Explore the repo to understand the current state. Sketch the major modules to build or modify. Look for opportunities to extract deep modules testable in isolation.

**If the conversation already contains scouted details (file paths, API formats, exact commands) — don't re-scout. Use what you have.**

## 2. Write the PRD

Save as `plan/NNNN-prd-....md` in the active repo root (the directory pi is running in, not a sub-project). Frontmatter:
```yaml
---
id: NNNN
title: PRD: ...
kind: prd
status: needs-triage
blocked_by: []
created: YYYY-MM-DD
---
```

Sections:
```markdown
## Problem Statement
The problem from the user's perspective.

## Solution
The solution from the user's perspective.

## User Stories
Numbered list (As an <actor>, I want <feature>, so that <benefit>).

## Implementation Decisions
- Modules built/modified
- Interfaces modified
- Architectural decisions
- API contracts
- If concrete details exist (file paths, endpoints, commands): include them. The PRD is the canonical record.

## Testing Decisions
- What makes a good test here
- Which modules will be tested

## Out of Scope
```

**If the conversation has a concrete plan** (exact files, APIs, Nix expressions): preserve those details. The PRD is the canonical record. Don't strip specifics just because a template says "abstract."

**If still at the problem/idea stage**: keep it abstract. No file paths.

## 3. Ask About Breakdown

Use `ask_user_decisions`:
- Q1: "Break this PRD into tracking issues now?" (Yes / No, let me review first)
- Q2: "How granular?" (Coarse milestones / Granular vertical slices / Extremely granular commits)

If "No", stop. If "Yes", proceed.

## 4. Break into Vertical Slices

Draft vertical slices (tracer bullets) that cut through ALL layers end-to-end. Do not slice by layer. Each slice must be deliverable on its own.

Classify each:
- **AFK**: Can be implemented and merged without human interaction.
- **HITL**: Needs human-in-the-loop decisions.

Quiz the user on titles, type, blockers, user stories covered. Wait for approval.

## 5. Create Tracking Issues

For each approved slice, create `plan/NNNN-....md` in the active repo root, in dependency order. Frontmatter:
```yaml
---
id: NNNN
title: ...
kind: enhancement
status: ready-for-agent
slice_type: AFK
blocked_by: []
created: YYYY-MM-DD
---
```

Sections:
```markdown
## Parent
`plan/NNNN-prd-....md`

## What to build
Concise description of this vertical slice end-to-end.

## Acceptance criteria
- [ ] Criterion

## Blocked by
None (or list issue IDs)
```

If concrete details exist in the PRD, include them in each issue — the agent picking up the issue should have everything needed without re-reading the PRD.

From drlove2002/ai-config, MIT.

More from drlove2002/ai-config

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jul 17, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

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

More from drlove2002

Cite this page

plan skill. Agent Mart. https://agentmart.dev/skills/drlove2002-plan (last updated 2026-10-24)