Memory Search

# Memory Search

You have two tools for recalling information from your memory files. Use them.

## Tools

### `memory_search`

Semantic vector search across your indexed memory files (MEMORY.md, memory/*.md, and session transcripts).

**Parameters:**

| Param | Type | Required | Description |
|---|---|---|---|
| `query` | string | yes | Natural language question or topic to search for |
| `maxResults` | number | no | Max results to return (default: 6) |
| `minScore` | number | no | Minimum relevance score threshold (0-1) |

**Example calls:**

```json
{ "query": "what projects is the human working on" }
{ "query": "preferences about code style", "maxResults": 3 }
{ "query": "important dates birthdays deadlines", "maxResults": 10, "minScore": 0.3 }
```

**Returns:** Array of results, each with:
- `snippet` — the matching text chunk
- `path` — relative file path (e.g. `MEMORY.md`, `memory/2026-02-07.md`)
- `startLine` / `endLine` — line range in the source file
- `score` — relevance score
- `citation` — formatted source reference (in direct chats)

### `memory_get`

Read a specific section of a memory file by path and line range. Use this after `memory_search` to pull more context around a result.

**Parameters:**

| Param | Type | Required | Description |
|---|---|---|---|
| `path` | string | yes | Relative path from workspace (e.g. `MEMORY.md`, `memory/2026-02-07.md`) |
| `from` | number | no | Starting line number |
| `lines` | number | no | Number of lines to read |

**Example calls:**

```json
{ "path": "MEMORY.md" }
{ "path": "memory/2026-02-07.md", "from": 15, "lines": 30 }
```

## When to Use Memory Search

**Always search before answering about:**

- Prior conversations or decisions
- The human's preferences, habits, or opinions
- Dates, deadlines, birthdays, events
- Project status or history
- Anything the human said "remember this" about
- Todos, action items, or commitments
- People, names, relationships

**The pattern is:**

1. Receive a question that m...