Skip to main content
skill CX Codex

02-user-stories — Claude Code skill: what it does, how to install

02-user-stories turns an epic, product brief or PRD into ordered user stories with acceptance criteria, then checks whether they are ready to build.

MIT ★ 571 updated 2026-10-29

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add ai-driven-dev/framework
/plugin install aidd-pm@aidd-framework
Open on GitHub

What 02-user-stories does

Give it a brief, a PRD, an epic or a rough request and it moves in stages: pin down the source and scope, cut vertical slices, draft stories with acceptance criteria, test readiness, then estimate and order where that helps. Only the file for the next stage gets opened, so the context stays tight.

The call takes request or epic, set through argument-hint. Product and lifecycle calls stay with you. Nothing lands on disk without your explicit approval, and the draft waits in the chat until you give it. Source links and edits you already made are kept.

Epics themselves and implementation work are out of scope.

What it helps with

  • slice an epic into vertical stories with acceptance criteria
  • check story readiness before development starts
  • order a backlog by estimate and dependency
  • turn a vague brief into concrete scenarios
  • persist an approved set of stories to a file

Example prompts for the agent

  • Slice the onboarding epic into stories with acceptance criteria and order them.
  • Check whether these stories are ready and name the blockers.
  • Persist the approved stories and keep the links back to the PRD.

How to install 02-user-stories

1Plugin marketplace

/plugin marketplace add ai-driven-dev/framework
/plugin install aidd-pm@aidd-framework

This installs the whole aidd-pm plugin with all its skills

2Skills CLI (Vercel)

npx skills add ai-driven-dev/framework --skill 02-user-stories -a claude-code

3Manually: put the files in place

~/.claude/skills/02-user-stories/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/02-user-stories/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about 02-user-stories

How do I install 02-user-stories in Claude Code?

Run npx skills add ai-driven-dev/framework --skill 02-user-stories -a claude-code and the skill file appears at ~/.claude/skills/02-user-stories/SKILL.md.

What does 02-user-stories do?

It takes an epic, product brief, PRD or bounded request and produces ordered user stories with acceptance criteria. Readiness checks, estimation and ordering come along the way, and saving happens once you approve.

Does it write stories to disk on its own?

No. Its rules require explicit approval or authority handed over by the caller before any write, so the draft stays in the conversation until then.

Source text

SKILL.md · MIT
---
name: 02-user-stories
description: Produces or refines ordered User Stories from an Epic, Product Brief, PRD, or bounded request. Use when the user wants to slice, write, assess, order, or persist Stories. Not for Epics or implementation.
argument-hint: request | epic
---

# User Stories

```mermaid
flowchart LR
  source([request, Product Brief, PRD, Epic, or Stories]) --> frame --> slice --> write --> assess
  source -->|"already persisted"| finalize
  assess -->|"reshape"| slice
  assess --> estimation --> order --> finalize
  order -->|"reslice"| slice
  finalize -->|"revise"| write
  finalize -->|"authorized"| done([User Stories])
```

## Actions

Run the flow above. Read only the next action file.

| Action     | Does                                      |
| ---------- | ----------------------------------------- |
| frame      | resolve the source and Story scope        |
| slice      | find vertical deliverable slices          |
| write      | write Stories and acceptance              |
| assess     | determine readiness and blockers          |
| estimation | estimate only when applicable             |
| order      | order only when useful                    |
| finalize   | approve, persist, or transition           |

## Transversal rules

- Keep product and lifecycle decisions with the user.
- Separate evidence, decisions, and assumptions.
- Preserve source links and existing edits.
- Ask natural questions; never expose actions, references, or unchanged state.
- Require explicit approval or caller-provided bounded authority before any write.
- Draft only after actor, need, and outcome are explicit in the source or confirmed.

© AI-Driven Dev · MIT · ai-driven-dev/framework

Similar skills

All skills for Notes, tasks and projects →

From the same set80

The whole set →
01-commit The 01-commit skill stages changes one concern at a time, writes a conventional commit message, and commits locally. It can push the branch when asked. CX Codex +1 ★ 571repo stars 02-implement 02-implement implements an existing plan in code, phase by phase, until every acceptance criterion passes. Useful when the plan is ready and the code is not. CX Codex +1 ★ 571repo stars 04-issue-create 04-issue-create is a skill for Claude Code that prepares and opens issues in a configured project tracker. It asks for explicit approval before submitting a ticket. CX Codex +1 ★ 571repo stars 02-pull-request 02-pull-request collects changes from the current branch and opens a draft pull or merge request. Use it when you want a review request prepared without publishing it as ready. CX Codex +1 ★ 571repo stars release-video release-video plans scenes for newly shipped features and renders an announcement video with background music and UI audio. CX Codex +1 ★ 571repo stars 01-acceptance-qa 01-acceptance-qa is a Claude Code skill that checks browser-visible behavior against acceptance criteria and records evidence videos. It verifies changes that have already been reviewed. CX Codex +1 ★ 571repo stars 08-debug 08-debug — a bug-fixing skill with three routes: reproduce a known failure and patch it with a test, chase an unknown cause through hypotheses, or reopen a stalled hunt. CX Codex +1 ★ 571repo stars executor executor turns an approved task scope into working, validated code. Useful when a spec needs implementation without planning drift or self-judging. CX Codex +1 ★ 571repo stars

Project sponsors

Companies that keep this analytics open