Memory Initializer

Set up project memory structure for a new project.

The Job

•Determine project name and details
•Create memory directory structure
•Initialize context.json with project info
•Create empty sessions.json and decisions.json

Step 1: Gather Project Info

Ask the user or infer from codebase:

code

Setting up project memory. Please confirm:

1. Project name: {inferred from package.json or directory}
2. Tech stack: {inferred from dependencies}
3. Brief description: {from README or ask}

Use AskUserQuestion if needed to clarify.

Step 2: Create Directory Structure

bash

mkdir -p ~/aiconfig/memory/projects/{project-name}

Step 3: Initialize context.json

Use atomic writes to ensure file integrity:

bash

# Atomic write with JSON validation
~/aiconfig/scripts/atomic-write.sh ~/aiconfig/memory/projects/{project-name}/context.json

Content:

json

{
  "project": "{project-name}",
  "created": "{ISO date}",
  "description": "{brief description}",
  "repository": "{git remote URL if available}",
  "tech_stack": {
    "languages": ["TypeScript"],
    "frameworks": ["Next.js", "React"],
    "databases": ["PostgreSQL"],
    "infrastructure": ["Vercel"]
  },
  "architecture": {
    "pattern": "{monolith|microservices|serverless}",
    "key_directories": {
      "src/components": "React components",
      "src/lib": "Shared utilities",
      "src/app": "Next.js app router"
    }
  },
  "current_focus": "",
  "active_branches": [],
  "known_issues": [],
  "team_conventions": {
    "commit_format": "conventional",
    "branch_naming": "type/description",
    "pr_template": true
  }
}

Step 4: Initialize sessions.json

json

{
  "project": "{project-name}",
  "sessions": []
}

Step 5: Initialize decisions.json

json

{
  "project": "{project-name}",
  "decisions": []
}

Step 6: Scan for Existing Decisions

Look for existing architectural decisions in:

•docs/adr/ or docs/decisions/
•ARCHITECTURE.md
•README.md technical sections

If found, offer to import them:

code

Found existing ADRs in docs/adr/. Import to memory?

Output

code

Project memory initialized: ~/aiconfig/memory/projects/{project-name}/

Created:
  - context.json (project overview and tech stack)
  - sessions.json (session history - empty)
  - decisions.json (architectural decisions - empty)

Next steps:
1. Review context.json and add any missing details
2. Use /log-session at end of coding sessions
3. Memory will be available in future sessions

Checklist

• Confirmed project name with user
• Created memory directory
• Populated context.json with tech stack
• Created empty sessions.json
• Created empty decisions.json
• Checked for existing ADRs to import