team-shinchan:memories

Memories Skill

View what Team-Shinchan has learned about this project.

Usage

/team-shinchan:memories              # View all learnings
/team-shinchan:memories pattern      # Filter by category
/team-shinchan:memories search auth  # Search for keyword

Process

Step 1: Read Learning File

Read .shinchan-docs/learnings.md

If file doesn't exist:

📚 [Memories] No learnings yet.

Start teaching with:
  /team-shinchan:learn "Your learning here"

Or just work - I learn automatically after tasks!

Step 2: Parse and Display

Full view (/memories):

📚 [Memories] {N} learnings stored

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🔹 PREFERENCE TIER ({count}) — always loaded
• Use Zustand over Redux [high] #state-management
• Prefer functional components [high] #react

🔸 PROCEDURAL TIER ({count}) — project-scoped
  📌 PATTERNS ({count})
  • Early returns for validation [high] #typescript
  • Repository pattern for data access [medium] #architecture
  📌 CONVENTIONS ({count})
  • Use pnpm, not npm [high] #tooling
  • camelCase for functions [high] #naming
  📌 MISTAKES ({count})
  • Always null-check before .map() [high] #typescript
  📌 DECISIONS ({count})
  • JWT with refresh token rotation [high] #auth

🔧 TOOL TIER ({count}) — execution-stage only
• Always pass --json to gh CLI [high] #github

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

💡 These learnings are applied automatically.
🗑️ Use /team-shinchan:forget to remove outdated ones.

Filter by tier (/memories --tier preference): Show only entries matching the specified tier.

Filtered view (/memories pattern):

📚 [Memories] Showing: patterns

• Early returns for validation [high]
  Context: Learned from UserService refactor

• Repository pattern for data access [medium]
  Context: Applied during database migration

Found 2 patterns.

Search view (/memories search auth):

📚 [Memories] Search: "auth"

1. [decision] JWT with refresh token rotation
   Tags: #auth #security #jwt

2. [pattern] Auth middleware pattern
   Tags: #auth #middleware

Found 2 results.

Memory Categories

| Category | Description | Icon | |----------|-------------|------| | preference | User preferences | 💜 | | pattern | Reusable code patterns | 🔷 | | convention | Project conventions | 📏 | | mistake | Errors to avoid | ⚠️ | | decision | Architecture decisions | 🏗️ | | insight | General insights | 💡 |

Important

• Learnings are stored in .shinchan-docs/learnings.md
• Automatically loaded at session start
• Higher confidence learnings are prioritized
• Old learnings can be removed with /forget