Back to skills
SKILL.md
Memory Memory Search
ASecuritySearch stored memory using semantic, keyword, or hybrid search.
- 21 stars
- 0 votes
- 0 copies
- 2 views
- Added September 10, 2026
Works with
Security analysis
100/100npx -y skills add monoes/monomind --skill memory-memory-search --agent claude-codeAre you the author of Memory Memory Search?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/monoes-memory-memory-search)---
name: memory-memory-search
description: Search stored memory using semantic, keyword, or hybrid search.
type: flow
---
# memory search
Search stored memory using semantic, keyword, or hybrid search.
## Usage
```bash
npx monomind memory search [options]
```
## Options
| Flag | Short | Type | Default | Description |
|---|---|---|---|---|
| `--query` | `-q` | string | — | Search query text |
| `--namespace` | `-n` | string | `default` | Namespace to search |
| `--limit` | `-l` | number | `10` | Maximum results to return |
| `--threshold` | — | number | `0.7` | Minimum similarity threshold (0–1) |
| `--type` | `-t` | string | `hybrid` | Search type: `semantic`, `keyword`, `hybrid` |
| `--build-hnsw` | — | boolean | `false` | Build pure-JS HNSW index before search (O(log n) vs O(n) linear) |
| `--format` | — | string | — | Output format: `json` |
## Examples
```bash
# Semantic search
npx monomind memory search --query "authentication patterns"
# Keyword search
npx monomind memory search -q "JWT" --type keyword
# Hybrid search with namespace
npx monomind memory search -q "error handling" -n "project-x" --type hybrid
# Build HNSW index first for large datasets
npx monomind memory search -q "config" --build-hnsw --limit 20
# High-precision search with threshold
npx monomind memory search -q "OAuth flow" --threshold 0.9
# JSON output
npx monomind memory search -q "auth" --format json
```
## Search Types
| Type | Description | Best For |
|---|---|---|
| `semantic` | Vector similarity search | Concept/meaning matching |
| `keyword` | BM25 full-text search | Exact term matching |
| `hybrid` | Combined semantic + keyword | General purpose (recommended) |
## HNSW Index
Use `--build-hnsw` when searching large memory stores (1,000+ entries). The pure-JS HNSW index gives O(log n) approximate-nearest-neighbor queries (vs O(n) linear scan) after the one-time build cost.
## MCP Tool
```javascript
mcp__monomind__memory_pattern-search({
query: "authentication patterns",
namespace: "default",
limit: 10,
threshold: 0.7,
type: "hybrid"
})
```
## See Also
- `memory store` — store new memories
- `memory list` — list all memories
- `memory retrieve` — retrieve by exact key
Attribution
Comments
Loading comments…