setup-project

name	setup-project
description	Initialize a freshly copied Cortex single-project template by detecting the tech stack and populating AGENTS.md from your code. Run once after copying.
disable-model-invocation	true

Setup Project

Initialize a freshly copied Cortex single-project template. Run this command after copying the template files into your project.

Usage: /setup-project

---

STEP 1: ANALYZE PROJECT

1.1 Detect Tech Stack

Scan the project root and check for:

1. Detect tech stack by checking for:

   • package.json → Node.js project (check for framework: Next.js, React, Vue, Angular, etc.)
   • pubspec.yaml → Flutter/Dart project
   • requirements.txt or pyproject.toml or setup.py → Python project
   • go.mod → Go project
   • Cargo.toml → Rust project
   • pom.xml or build.gradle → Java/Kotlin project
   • Gemfile → Ruby project
   • .csproj or .sln → .NET project
   • appsscript.json → Google Apps Script
   • If none detected → Generic project

2. Analyze project structure:

   • Run ls on the project root to understand folder organization
   • Identify key directories (src/, lib/, app/, pages/, etc.)
   • Check for existing source code patterns
   • Note testing frameworks, linting tools, build tools

3. Identify key dependencies from package manager files

4. Detect build/test/lint commands from:

   • package.json scripts
   • Makefile targets
   • CI configuration files
   • README instructions

---

STEP 2: POPULATE AGENTS.md

Read the current AGENTS.md and fill in the skeleton sections. (The thin CLAUDE.md just imports AGENTS.md — do not populate it.)

2.1 Tech Stack Table

Fill in the Tech Stack table with detected framework, language, package manager, testing framework, and database.

2.2 Commands Section

Fill in the Commands section with detected build, test, lint, and dev server commands.

2.3 File Organization

Fill in the File Organization section with the actual project directory structure.

2.4 Coding Conventions

Analyze source code to detect and document:

• Naming conventions (camelCase, snake_case, etc.)
• Import patterns (relative vs absolute, barrel files, etc.)
• Error handling patterns
• State management approach (if applicable)

2.5 Important Patterns

Document 2-3 key recurring patterns found in the codebase.

2.6 Architecture Section

Fill in the Architecture section:

• Architecture style (monolith, microservices, etc.)
• Key directories and their purposes
• Data model: if discoverable, document key data models, schemas, or types
• Key flows: document 2-3 key data flows or processes in the application
• API structure: if applicable, document the API organization

---

STEP 3: INITIALIZE KNOWLEDGE BASE

3.1 Run Knowledge Base Population

Execute the logic from /populate-knowledge-base to:

• Scan the project for concepts, patterns, and flows
• Create initial topic files in the appropriate knowledge-base sections
• Generate INDEX.md and all _SUMMARY.md files

---

STEP 4: UPDATE STATUS

Update context-engineering/_STATUS.md with:

• Detected tech stack
• Initial project state (0 active PRPs, 0 features)
• Timestamp of project setup

---

STEP 5: REPORT

Present a summary report to the user:

=== Cortex Project Setup Complete ===

Project: {project-name}
Tech Stack: {detected tech stack}
Framework: {framework and version}

Documentation Updated:
  - AGENTS.md — Coding standards, file organization, commands, architecture, data model, key flows populated

Knowledge Base:
  - {N} initial topics created
  - INDEX.md generated

Next Steps:
  1. Review AGENTS.md and customize the coding standards and architecture details
  2. Start your first feature with /generate-requirements

---

SAFETY RULES

• NEVER modify project source code — this command only creates/updates documentation files
• NEVER commit to git — the user handles all git operations
• Read before writing — always read existing file content before updating
• Preserve user customizations — if a section already has user content, do not overwrite it