pm-workflow-copilot-ide

AI Product‑Management copilot for Cursor & Cline IDEs ‑ guiding you from idea ➜ MVP docs with chat‑first workflows.

---

💡 Example Ideas for Testing

This pack includes a memory_starters/examples directory containing several sample product ideas (e.g., weather.md, to-do_list.md).

To quickly test the product management workflow, you can start a conversation with your AI assistant about one of these ideas. For example, try saying:

"Let's start working on a weather app."

The assistant will then guide you through the process, creating documents based on the workflow defined in this pack.

---

🚀 Why this project?

Product teams at fast‑moving startups rarely have time to open Confluence. Yet skipping structured discovery & scoping leads to feature creep and re‑work. pm‑workflow‑copilot‑ide embeds a lightweight, opinionated product‑management workflow inside the editor chat you already use, keeping momentum while safeguarding first‑principles rigor.

---

🧩 How it works

1. Rules file is loaded by Cursor/Cline so the assistant understands the workflow and file‑naming conventions.
2. On first mention of a project (e.g. “weather app”) the copilot creates pm_project_docs/<project_name>/.
3. For every workflow step it suggests the next best action and offers to auto‑generate artefacts (Opportunity Brief, MVP Scope, Usability Findings…).
4. All artefacts are Markdown stored alongside code, enabling docs‑as‑code review flows.
5. Progress is logged to pm-progress.json for instant context recall.

---

💬 What Using the Copilot Feels Like (Sample Dialogs)

Below are a handful of concise interactions that illustrate typical usage. Real wording varies by IDE, but the flow stays the same.

1. Kick‑off from a Rough Idea

You: I want to build an app that helps freelancers track billable hours.
Copilot: Excellent! Let’s begin with Strategic Alignment. Who's the primary user and what pain are we solving?

2. Checking Project Progress

You: How far along is my "weather_app" project and what's next?
Copilot: Progress → Product Charter ✓  •  User Flow ✓  •  MVP Scope ✓  
Next → Prototype & usability testing → `usability_findings.md` → business viability check.

3. Clarifying Workflow Stage

You: Which phase are we in now?
Copilot: You're in Feasibility & Viability. Technical feasibility completed; business viability still pending. Shall I draft `business_viability.md`?

4. Requesting a New Artefact

You: Generate the MVP scope for the budgeting tool, please.
Copilot: Drafting `pm_project_docs/budgeting_tool/mvp_scope.md` with user stories, success metrics, and out‑of‑scope items…

(Conversation styles differ across IDEs; snippets are simplified for readability.)

---

📦 Installation

This pack can be used with an AI-powered IDE in two ways: automatically with the rulebook-ai CLI, or by manually configuring your IDE.

With Rulebook-AI (Recommended)

Rulebook-AI is a cross-IDE toolkit that manages AI assistant rules and memory.

1. Install the rulebook-ai CLI:

   First, clone the Rulebook-AI repository and install it in editable mode using uv:

   git clone https://github.com/botingw/rulebook-ai.git
   cd rulebook-ai
   uv run pip install -e .
   cd ..

2. Add the Pack:

   This repository contains the pm-workflow-copilot pack. From the root of the pm-workflow-copilot-ide project directory, or root of any repo you want install this pack, run the following command to add the pack:

   rulebook-ai packs add pm-workflow-copilot
   rulebook-ai project sync --pack pm-workflow-copilot

   The CLI will validate the pack and make it available to your IDE's AI assistant.

3. Start Chatting:

   You're all set. Open a new chat and start with an idea, like "Let's work on a weather app."

Tip: keep your product‑management artefacts in git. They get versioned automatically, and reviewers can PR suggestions like any code change.

---

📦 Pack Layout

This pack has a specific structure that rulebook-ai recognizes. Here is an overview of the key directories and files within the pm-workflow-copilot-pack and what they do.

pm-workflow-copilot-pack/
├── manifest.yaml
├── README.md
├── memory_starters/
│   ├── docs/
│   │   ├── product_management_workflow_startup.md
│   │   └── pm-progress-template.json
│   └── examples/
│       └── ...
└── rules/
    └── 01-rules/
        └── 01-pm_workflow_assistant.md

Key Components

• manifest.yaml: This file contains metadata about the pack, such as its name (pm-workflow-copilot), version, and a brief summary. The rulebook-ai CLI uses this for identification.

• rules/: This is the core of the pack. It contains the instructions that tell the AI assistant how to behave.

  • 01-pm_workflow_assistant.md: The main rule file. It defines the entire product management workflow, from strategic alignment to final requirements. It tells the assistant how to guide the user, what questions to ask, and which documents to generate.

• memory_starters/: This directory contains documents that are loaded into the AI's short-term memory when the pack is activated. This gives the assistant the foundational knowledge it needs to operate.

  • docs/product_management_workflow_startup.md: The detailed, formal specification of the product management workflow. The assistant consults this as its primary reference.
  • docs/pm-progress-template.json: The JSON schema for the progress tracking file.
  • examples/: Sample product ideas that you can use to test the pack's workflow.

Generated Project Artifacts

When you use this pack in your own project, the assistant will generate a new directory at the root of your project folder:

• pm_project_docs/: This folder is created automatically in your project's root directory the first time you start working on a new product idea with the assistant.
  • [project_name]/: Inside pm_project_docs, a sub-directory is created for each product idea you work on (e.g., weather_app/).
  • *.md: All the product management documents (Product Charter, PRD, etc.) are saved as Markdown files inside their respective project folder.
  • pm-progress.json: A file that tracks your progress through the workflow for a specific project.

This structure keeps all your product documentation neatly organized alongside your code, right where you work.

---

📓 Workflow Cheatsheet

Phase	Key Output	Trigger Command / Prompt
Strategic Alignment	product_charter.md	"Who is this for?"
Problem Discovery	opportunity_brief.md	"Draft 5 survey Qs"
Solution Definition	mvp_scope.md	"Prioritize MVP features"
Feasibility & Viability	technical_feasibility.md business_viability.md	"Assess tech risks"
Requirements	prd.md	"Generate PRD"

---

⚠️ Experimental Status

Heads‑up: pm‑workflow‑copilot‑ide began as a proving ground for bringing lightweight product‑management support into the broader Rulebook‑AI ecosystem — a cross‑IDE "rules + memory bank" toolkit for AI coding assistants. The long‑term plan is to merge this functionality upstream and deprecate this standalone repo once parity is reached. Follow the Rulebook‑AI repo for consolidated updates.

---

🗺️ Roadmap

• Context‑aware commands for VS Code extension
• Slack bot companion for cross‑functional reviews
• Automatic link‑back between PRs and related PM artefacts

Check the issues list for more.

---

🤝 Contributing

1. Fork and create a feature branch.
2. Follow Conventional Commits for commit messages.
3. Run pre‑commit install to auto‑format Markdown.
4. Open a PR – the README’s badges will update once CI passes.

We love fresh perspectives from PMs, designers & engineers alike.

---

🔒 License

Distributed under the MIT License. See LICENSE for details.

---

🙏 Acknowledgements

• Inspired by Tom Preston‑Werner’s '''README first''' philosophy.
• Workflow adapted from the open‑source [Product Management Workflow for Startups].
• Badges by shields.io.# pm-workflow-copilot-ide