Generating Doc Packs

The "Doc Pack" is the final output of Context Ark. It acts as a comprehensive "Project Bible" that you can hand off to human engineers or AI agents.

The Folder Structure

The 60+ documents are organized into a strict hierarchy:

1/docs
2  ├── 01_Product_Strategy
3  │   ├── 01_PRD.md
4  │   ├── 02_USER_STORIES.md
5  │   └── ...
6  ├── 02_User_Experience
7  │   ├── 01_PERSONAS.md
8  │   ├── 02_USER_FLOWS.md
9  │   └── ...
10  ├── 03_Context_Engineering (CRITICAL)
11  │   ├── CURSOR_RULES.md
12  │   ├── TECH_STACK_RULES.md
13  │   └── ...
14  ├── 04_System_Architecture
15  │   ├── 01_SYSTEM_OVERVIEW.md
16  │   ├── 02_DATABASE_SCHEMA.md
17  │   └── ...
18  └── 05_Security_and_Ops
19      ├── 01_SECURITY_POLICY.md
20      └── ...

Key Files Explained

CURSOR_RULES.md

This is arguably the most valuable file. It contains specific instructions for the Cursor IDE. Copy the contents of this file into your .cursorrules file at the root of your project to give the AI editor full context of your project's standards.

DATABASE_SCHEMA.md

Contains a mermaid ER diagram and SQL definitions. You can paste this directly into a tool like v0 or Phind to generate your migration files.

01_PRD.md

The Product Requirement Document. It serves as the "contract" for what the application must do.

Regenerating Individual Files

If you change a small detail (e.g., adding a feature), you don't always need to regenerate the entire pack (which costs 50 credits).

1. Navigate to the Docs tab in your project.
2. Find the specific file (e.g., 01_PRD.md).
3. Click the Regenerate (Refresh icon) button.
4. Cost: 1 Credit.

Downloading

You can download the entire pack as a ZIP file. The internal links between documents (e.g., [Link](./01_PRD.md)) are preserved, so it works perfectly as a local knowledge base in Obsidian or VS Code.