# Cookbook API Reference
---
name: "Cookbook API Reference"
description: "Complete cookbook() tool API documentation for Vector Memory MCP"
part: 1
type: "api"
date: "2026-02-19"
version: "2.0.0"
---
## Parameters
```python
mcp__vector-memory__cookbook(
include: str = "init", # What to return
level: int = 0, # Docs verbosity (0-3)
case_category: str = None, # Filter by category key(s), CSV for OR
query: str = None, # Text search
priority: str = None, # Filter by priority (critical, high, or both)
cognitive: str = None, # Filter by cognitive level tags
strict: str = None, # Filter by strict level tags
limit: int = 10, # Max results (1-50)
offset: int = 0 # Pagination offset
)
```
## Parameter Details
### include
| Value | Description |
|-------|-------------|
| `init` | FIRST READ - quick start + available resources |
| `docs` | Documentation by level |
| `cases` | Use case scenarios (filtered) |
| `categories` | List categories with keys |
| `all` | Everything combined (large!) |
### level
| Value | Content |
|-------|---------|
| 0 | Identity & Quick Start |
| 1 | Practical Usage |
| 2 | Advanced Patterns |
| 3 | Architecture & Internals |
### case_category
Single key or comma-separated list (OR logic):
```python
# Single
case_category="gates-rules"
# Multiple (OR)
case_category="store,search,gates-rules"
```
### query
Text search across all content:
```python
query="JWT token"
query="multi-probe search"
query="CRITICAL"
```
### priority
Filter by priority markers in content:
| Value | Description |
|-------|-------------|
| `critical` | Only [CRITICAL] content |
| `high` | Only [HIGH] content |
| `critical,high` | Both CRITICAL and HIGH |
### cognitive
Filter by `cognitive:*` tags in content (OR logic):
| Value | Description |
|-------|-------------|
| `minimal` | cognitive:minimal patterns |
| `standard` | cognitive:standard patterns |
| `deep` | cognitive:deep patterns |
| `exhaustive` | cognitive:exhaustive patterns |
| `deep,exhaustive` | Both deep and exhaustive |
### strict
Filter by `strict:*` tags in content (OR logic):
| Value | Description |
|-------|-------------|
| `relaxed` | strict:relaxed patterns |
| `standard` | strict:standard patterns |
| `strict` | strict:strict patterns |
| `paranoid` | strict:paranoid patterns |
| `strict,paranoid` | Both strict and paranoid |
### limit / offset
Pagination for large results:
```python
limit=10, offset=0 # First page
limit=10, offset=10 # Second page
```
## Examples
### Initialization
```python
mcp__vector-memory__cookbook()
mcp__vector-memory__cookbook(include="init")
```
### Categories
```python
mcp__vector-memory__cookbook(include="categories")
```
### Single Category
```python
mcp__vector-memory__cookbook(include="cases", case_category="gates-rules")
```
### Multiple Categories (OR)
```python
mcp__vector-memory__cookbook(include="cases", case_category="store,search")
```
### Priority Filter
```python
# Only CRITICAL
mcp__vector-memory__cookbook(include="cases", priority="critical")
# CRITICAL or HIGH
mcp__vector-memory__cookbook(include="cases", priority="critical,high")
```
### Cognitive Filter
```python
# Only deep
mcp__vector-memory__cookbook(include="cases", cognitive="deep")
# Deep or exhaustive
mcp__vector-memory__cookbook(include="cases", cognitive="deep,exhaustive")
```
### Strict Filter
```python
# Only strict
mcp__vector-memory__cookbook(include="cases", strict="strict")
# Strict or paranoid
mcp__vector-memory__cookbook(include="cases", strict="strict,paranoid")
```
### Search
```python
mcp__vector-memory__cookbook(include="cases", query="JWT")
mcp__vector-memory__cookbook(include="docs", query="tag normalization", level=2)
```
### Combined Filters
```python
mcp__vector-memory__cookbook(
include="cases",
case_category="gates-rules",
query="secret",
priority="critical",
limit=5
)
# Brain with cognitive:deep, strict:standard
mcp__vector-memory__cookbook(
include="cases",
cognitive="deep,exhaustive",
strict="standard,strict,paranoid",
limit=20
)
```
### Pagination
```python
mcp__vector-memory__cookbook(include="cases", priority="critical", limit=5, offset=0)
mcp__vector-memory__cookbook(include="cases", priority="critical", limit=5, offset=5)
```
## Consistency with Task MCP
Both MCPs use IDENTICAL cookbook() parameters:
| Parameter | Memory MCP | Task MCP |
|-----------|------------|----------|
| include | ✓ same | ✓ same |
| level | ✓ same | ✓ same |
| case_category | ✓ same | ✓ same |
| query | ✓ same | ✓ same |
| priority | ✓ same | ✓ same |
| cognitive | ✓ same | ✓ same |
| strict | ✓ same | ✓ same |
| limit | ✓ same | ✓ same |
| offset | ✓ same | ✓ same |
This ensures agents can use cookbook() consistently across both MCPs.
## Response Structure
### init
```json
{
"success": true,
"critical": "READ THIS FIRST",
"warning": "Without this cookbook...",
"quick_start": {...},
"available_resources": {
"cookbook_docs": {...},
"cookbook_cases": {...},
"cookbook_priority": {...},
"cookbook_cognitive": {...},
"cookbook_strict": {...},
"cookbook_search": {...},
"cookbook_categories": {...}
},
"message": "Retrieved init level 0"
}
```
### cases
```json
{
"success": true,
"include": "cases",
"level": 0,
"case_category": "gates-rules",
"query": null,
"priority": "critical",
"cognitive": null,
"strict": null,
"limit": 10,
"offset": 0,
"cases": {
"categories": ["Gates & Rules Scenarios"],
"count": 1,
"content": "..."
},
"message": "Retrieved cases level 0 priority='critical'"
}
```
