agentdb-advanced

Name: agentdb-advanced
Author: spencermarx

301stars24forks

How to install agentdb-advanced?

spencermarx/open-code-review/agentdb-advanced

$npx -y skills add spencermarx/open-code-review --skill agentdb-advanced

Installs into the current project.

Summary

Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid search, and distributed systems integration. Use when building distributed AI systems, multi-agent coordination, or advanced vector search applications.

Files · 1

View on GitHub

SKILL.md

1	# AgentDB Advanced Features
2
3	## What This Skill Does
4
5	Covers advanced AgentDB capabilities for distributed systems, multi-database coordination, custom distance metrics, hybrid search (vector + metadata), QUIC synchronization, and production deployment patterns. Enables building sophisticated AI systems with sub-millisecond cross-node communication and advanced search capabilities.
6
7	Performance: <1ms QUIC sync, hybrid search with filters, custom distance metrics.
8
9	## Prerequisites
10
11	- Node.js 18+
12	- AgentDB v1.0.7+ (via agentic-flow)
13	- Understanding of distributed systems (for QUIC sync)
14	- Vector search fundamentals
15
16	---
17
18	## QUIC Synchronization
19
20	### What is QUIC Sync?
21
22	QUIC (Quick UDP Internet Connections) enables sub-millisecond latency synchronization between AgentDB instances across network boundaries with automatic retry, multiplexing, and encryption.
23
24	Benefits:
25	- <1ms latency between nodes
26	- Multiplexed streams (multiple operations simultaneously)
27	- Built-in encryption (TLS 1.3)
28	- Automatic retry and recovery
29	- Event-based broadcasting
30
31	### Enable QUIC Sync
32
33	```typescript
34	import { createAgentDBAdapter } from 'agentic-flow/reasoningbank';
35
36	// Initialize with QUIC synchronization
37	const adapter = await createAgentDBAdapter({
38	dbPath: '.agentdb/distributed.db',
39	enableQUICSync: true,
40	syncPort: 4433,
41	syncPeers: [
42	'192.168.1.10:4433',
43	'192.168.1.11:4433',
44	'192.168.1.12:4433',
45	],
46	});
47
48	// Patterns automatically sync across all peers
49	await adapter.insertPattern({
50	// ... pattern data
51	});
52
53	// Available on all peers within ~1ms
54	```
55
56	### QUIC Configuration
57
58	```typescript
59	const adapter = await createAgentDBAdapter({
60	enableQUICSync: true,
61	syncPort: 4433, // QUIC server port
62	syncPeers: ['host1:4433'], // Peer addresses
63	syncInterval: 1000, // Sync interval (ms)
64	syncBatchSize: 100, // Patterns per batch
65	maxRetries: 3, // Retry failed syncs
66	compression: true, // Enable compression
67	});
68	```
69
70	### Multi-Node Deployment
71
72	```bash
73	# Node 1 (192.168.1.10)
74	AGENTDB_QUIC_SYNC=true \
75	AGENTDB_QUIC_PORT=4433 \
76	AGENTDB_QUIC_PEERS=192.168.1.11:4433,192.168.1.12:4433 \
77	node server.js
78
79	# Node 2 (192.168.1.11)
80	AGENTDB_QUIC_SYNC=true \
81	AGENTDB_QUIC_PORT=4433 \
82	AGENTDB_QUIC_PEERS=192.168.1.10:4433,192.168.1.12:4433 \
83	node server.js
84
85	# Node 3 (192.168.1.12)
86	AGENTDB_QUIC_SYNC=true \
87	AGENTDB_QUIC_PORT=4433 \
88	AGENTDB_QUIC_PEERS=192.168.1.10:4433,192.168.1.11:4433 \
89	node server.js
90	```
91
92	---
93
94	## Distance Metrics
95
96	### Cosine Similarity (Default)
97
98	Best for normalized vectors, semantic similarity:
99
100	```bash
101	# CLI
102	npx agentdb@latest query ./vectors.db "[0.1,0.2,...]" -m cosine
103
104	# API
105	const result = await adapter.retrieveWithReasoning(queryEmbedding, {
106	metric: 'cosine',
107	k: 10,
108	});
109	```
110
111	Use Cases:
112	- Text embeddings (BERT, GPT, etc.)
113	- Semantic search
114	- Document similarity
115	- Most general-purpose applications
116
117	Formula: `cos(θ) = (A · B) / (\|\|A\|\| × \|\|B\|\|)`
118	Range: [-1, 1] (1 = identical, -1 = opposite)
119
120	### Euclidean Distance (L2)
121
122	Best for spatial data, geometric similarity:
123
124	```bash
125	# CLI
126	npx agentdb@latest query ./vectors.db "[0.1,0.2,...]" -m euclidean
127
128	# API
129	const result = await adapter.retrieveWithReasoning(queryEmbedding, {
130	metric: 'euclidean',
131	k: 10,
132	});
133	```
134
135	Use Cases:
136	- Image embeddings
137	- Spatial data
138	- Computer vision
139	- When vector magnitude matters
140
141	Formula: `d = √(Σ(ai - bi)²)`
142	Range: [0, ∞] (0 = identical, ∞ = very different)
143
144	### Dot Product
145
146	Best for pre-normalized vectors, fast computation:
147
148	```bash
149	# CLI
150	npx agentdb@latest query ./vectors.db "[0.1,0.2,...]" -m dot
151
152	# API
153	const result = await adapter.retrieveWithReasoning(queryEmbedding, {
154	metric: 'dot',
155	k: 10,
156	});
157	```
158
159	Use Cases:
160	- Pre-normalized embeddings
161	- Fast similarity computation
162	- When vectors are already unit-length
163
164	Formula: `dot = Σ(ai × bi)`
165	Range: [-∞, ∞] (higher = more similar)
166
167	### Custom Distance Metrics
168
169	```typescript
170	// Implement custom distance function
171	function customDistance(vec1: number[], vec2: number[]): number {
172	// Weighted Euclidean distance
173	const weights = [1.0, 2.0, 1.5, ...];
174	let sum = 0;
175	for (let i = 0; i < vec1.length; i++) {
176	sum += weights[i] * Math.pow(vec1[i] - vec2[i], 2);
177	}
178	return Math.sqrt(sum);
179	}
180
181	// Use in search (requires custom implementation)
182	```
183
184	---
185
186	## Hybrid Search (Vector + Metadata)
187
188	### Basic Hybrid Search
189
190	Combine vector similarity with metadata filtering:
191
192	```typescript
193	// Store documents with metadata
194	await adapter.insertPattern({
195	id: '',
196	type: 'document',
197	domain: 'research-papers',
198	pattern_data: JSON.stringify({
199	embedding: documentEmbeddi

People who liked this also viewed

6 picks

Type

agentdb-advanced

301stars24forks

View on GitHub

How to install agentdb-advanced?

spencermarx/open-code-review/agentdb-advanced

$npx -y skills add spencermarx/open-code-review --skill agentdb-advanced

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# AgentDB Advanced Features
2
3	## What This Skill Does
4
5	Covers advanced AgentDB capabilities for distributed systems, multi-database coordination, custom distance metrics, hybrid search (vector + metadata), QUIC synchronization, and production deployment patterns. Enables building sophisticated AI systems with sub-millisecond cross-node communication and advanced search capabilities.
6
7	Performance: <1ms QUIC sync, hybrid search with filters, custom distance metrics.
8
9	## Prerequisites
10
11	- Node.js 18+
12	- AgentDB v1.0.7+ (via agentic-flow)
13	- Understanding of distributed systems (for QUIC sync)
14	- Vector search fundamentals
15
16	---
17
18	## QUIC Synchronization
19
20	### What is QUIC Sync?
21
22	QUIC (Quick UDP Internet Connections) enables sub-millisecond latency synchronization between AgentDB instances across network boundaries with automatic retry, multiplexing, and encryption.
23
24	Benefits:
25	- <1ms latency between nodes
26	- Multiplexed streams (multiple operations simultaneously)
27	- Built-in encryption (TLS 1.3)
28	- Automatic retry and recovery
29	- Event-based broadcasting
30
31	### Enable QUIC Sync
32
33	```typescript
34	import { createAgentDBAdapter } from 'agentic-flow/reasoningbank';
35
36	// Initialize with QUIC synchronization
37	const adapter = await createAgentDBAdapter({
38	dbPath: '.agentdb/distributed.db',
39	enableQUICSync: true,
40	syncPort: 4433,
41	syncPeers: [
42	'192.168.1.10:4433',
43	'192.168.1.11:4433',
44	'192.168.1.12:4433',
45	],
46	});
47
48	// Patterns automatically sync across all peers
49	await adapter.insertPattern({
50	// ... pattern data
51	});
52
53	// Available on all peers within ~1ms
54	```
55
56	### QUIC Configuration
57
58	```typescript
59	const adapter = await createAgentDBAdapter({
60	enableQUICSync: true,
61	syncPort: 4433, // QUIC server port
62	syncPeers: ['host1:4433'], // Peer addresses
63	syncInterval: 1000, // Sync interval (ms)
64	syncBatchSize: 100, // Patterns per batch
65	maxRetries: 3, // Retry failed syncs
66	compression: true, // Enable compression
67	});
68	```
69
70	### Multi-Node Deployment
71
72	```bash
73	# Node 1 (192.168.1.10)
74	AGENTDB_QUIC_SYNC=true \
75	AGENTDB_QUIC_PORT=4433 \
76	AGENTDB_QUIC_PEERS=192.168.1.11:4433,192.168.1.12:4433 \
77	node server.js
78
79	# Node 2 (192.168.1.11)
80	AGENTDB_QUIC_SYNC=true \
81	AGENTDB_QUIC_PORT=4433 \
82	AGENTDB_QUIC_PEERS=192.168.1.10:4433,192.168.1.12:4433 \
83	node server.js
84
85	# Node 3 (192.168.1.12)
86	AGENTDB_QUIC_SYNC=true \
87	AGENTDB_QUIC_PORT=4433 \
88	AGENTDB_QUIC_PEERS=192.168.1.10:4433,192.168.1.11:4433 \
89	node server.js
90	```
91
92	---
93
94	## Distance Metrics
95
96	### Cosine Similarity (Default)
97
98	Best for normalized vectors, semantic similarity:
99
100	```bash
101	# CLI
102	npx agentdb@latest query ./vectors.db "[0.1,0.2,...]" -m cosine
103
104	# API
105	const result = await adapter.retrieveWithReasoning(queryEmbedding, {
106	metric: 'cosine',
107	k: 10,
108	});
109	```
110
111	Use Cases:
112	- Text embeddings (BERT, GPT, etc.)
113	- Semantic search
114	- Document similarity
115	- Most general-purpose applications
116
117	Formula: `cos(θ) = (A · B) / (\|\|A\|\| × \|\|B\|\|)`
118	Range: [-1, 1] (1 = identical, -1 = opposite)
119
120	### Euclidean Distance (L2)
121
122	Best for spatial data, geometric similarity:
123
124	```bash
125	# CLI
126	npx agentdb@latest query ./vectors.db "[0.1,0.2,...]" -m euclidean
127
128	# API
129	const result = await adapter.retrieveWithReasoning(queryEmbedding, {
130	metric: 'euclidean',
131	k: 10,
132	});
133	```
134
135	Use Cases:
136	- Image embeddings
137	- Spatial data
138	- Computer vision
139	- When vector magnitude matters
140
141	Formula: `d = √(Σ(ai - bi)²)`
142	Range: [0, ∞] (0 = identical, ∞ = very different)
143
144	### Dot Product
145
146	Best for pre-normalized vectors, fast computation:
147
148	```bash
149	# CLI
150	npx agentdb@latest query ./vectors.db "[0.1,0.2,...]" -m dot
151
152	# API
153	const result = await adapter.retrieveWithReasoning(queryEmbedding, {
154	metric: 'dot',
155	k: 10,
156	});
157	```
158
159	Use Cases:
160	- Pre-normalized embeddings
161	- Fast similarity computation
162	- When vectors are already unit-length
163
164	Formula: `dot = Σ(ai × bi)`
165	Range: [-∞, ∞] (higher = more similar)
166
167	### Custom Distance Metrics
168
169	```typescript
170	// Implement custom distance function
171	function customDistance(vec1: number[], vec2: number[]): number {
172	// Weighted Euclidean distance
173	const weights = [1.0, 2.0, 1.5, ...];
174	let sum = 0;
175	for (let i = 0; i < vec1.length; i++) {
176	sum += weights[i] * Math.pow(vec1[i] - vec2[i], 2);
177	}
178	return Math.sqrt(sum);
179	}
180
181	// Use in search (requires custom implementation)
182	```
183
184	---
185
186	## Hybrid Search (Vector + Metadata)
187
188	### Basic Hybrid Search
189
190	Combine vector similarity with metadata filtering:
191
192	```typescript
193	// Store documents with metadata
194	await adapter.insertPattern({
195	id: '',
196	type: 'document',
197	domain: 'research-papers',
198	pattern_data: JSON.stringify({
199	embedding: documentEmbeddi

People who liked this also viewed

6 picks

Type