memory-management

# Memory Management Skill

Use AutoMem to maintain persistent context across Claude Code sessions. This skill teaches the **3-Phase Memory Pattern**: Recall → Store → Summarize.

## Phase 1: SESSION START (Recall)

### Always Recall For

- Project context questions (architecture, tooling, deployment)
- Architecture discussions or decisions
- User preferences and code style
- Debugging issues (search for similar past problems)
- Refactoring (understand why current structure exists)
- Integration or API work (check past implementations)
- Performance optimization discussions

### Adaptive Recall Based on Context

- **Files open** → Recall memories tagged with those components
- **Error messages** → Search for similar error patterns
- **Multiple files** → Recall architectural decisions
- **PR/commit context** → Recall related feature implementations

### Skip Recall For

- Pure syntax questions ("How does Array.map work?")
- Trivial edits (typos, formatting, simple renames)
- Direct factual queries about current code
- File content requests that can be answered by reading

### Recall Examples

```javascript
// Basic project recall
mcp__memory__recall_memory({
  query: "project architecture decisions",
  tags: ["project-name"],
  limit: 5
})

// Debug similar errors
mcp__memory__recall_memory({
  query: "TypeError authentication timeout",
  tags: ["bug-fix"],
  time_query: "last 30 days",
  limit: 5
})

// Multi-hop reasoning (find related context)
mcp__memory__recall_memory({
  query: "Who worked on the auth system?",
  expand_entities: true,
  limit: 10
})

// Context-aware coding recall
mcp__memory__recall_memory({
  query: "error handling patterns",
  language: "typescript",
  context: "coding-style",
  context_types: ["Style", "Pattern"]
})
```

## Phase 2: DURING WORK (Store)

### What to Store with Importance Levels

| Type | Importance | When to Store |
|------|------------|---------------|
| **Decision** | 0.9 | Architecture, library choices, pattern decisions