cloudflare-vectorize

Name: cloudflare-vectorize
Author: jackspace

15stars7forks

How to install cloudflare-vectorize?

jackspace/claudeskillz/cloudflare-vectorize

$npx -y skills add jackspace/ClaudeSkillz --skill cloudflare-vectorize

Installs into the current project.

Summary

Complete knowledge domain for Cloudflare Vectorize - globally distributed vector database for building semantic search, RAG (Retrieval Augmented Generation), and AI-powered applications. Use when: creating vector indexes, inserting embeddings, querying vectors, implementing seman

Files · 1

View on GitHub

SKILL.md

1	# Cloudflare Vectorize
2
3	Complete implementation guide for Cloudflare Vectorize - a globally distributed vector database for building semantic search, RAG (Retrieval Augmented Generation), and AI-powered applications with Cloudflare Workers.
4
5	Status: Production Ready ✅
6	Last Updated: 2025-10-21
7	Dependencies: cloudflare-worker-base (for Worker setup), cloudflare-workers-ai (for embeddings)
8	Latest Versions: wrangler@4.43.0, @cloudflare/workers-types@4.20251014.0
9	Token Savings: ~65%
10	Errors Prevented: 8
11	Dev Time Saved: ~3 hours
12
13	## What This Skill Provides
14
15	### Core Capabilities
16	- ✅ Index Management: Create, configure, and manage vector indexes
17	- ✅ Vector Operations: Insert, upsert, query, delete, and list vectors
18	- ✅ Metadata Filtering: Advanced filtering with 10 metadata indexes per index
19	- ✅ Semantic Search: Find similar vectors using cosine, euclidean, or dot-product metrics
20	- ✅ RAG Patterns: Complete retrieval-augmented generation workflows
21	- ✅ Workers AI Integration: Native embedding generation with @cf/baai/bge-base-en-v1.5
22	- ✅ OpenAI Integration: Support for text-embedding-3-small/large models
23	- ✅ Document Processing: Text chunking and batch ingestion pipelines
24
25	### Templates Included
26	1. basic-search.ts - Simple vector search with Workers AI
27	2. rag-chat.ts - Full RAG chatbot with context retrieval
28	3. document-ingestion.ts - Document chunking and embedding pipeline
29	4. metadata-filtering.ts - Advanced filtering examples
30
31	## Critical Setup Rules
32
33	### ⚠️ MUST DO BEFORE INSERTING VECTORS
34	```bash
35	# 1. Create the index with FIXED dimensions and metric
36	npx wrangler vectorize create my-index \
37	--dimensions=768 \
38	--metric=cosine
39
40	# 2. Create metadata indexes IMMEDIATELY (before inserting vectors!)
41	npx wrangler vectorize create-metadata-index my-index \
42	--property-name=category \
43	--type=string
44
45	npx wrangler vectorize create-metadata-index my-index \
46	--property-name=timestamp \
47	--type=number
48	```
49
50	Why: Metadata indexes MUST exist before vectors are inserted. Vectors added before a metadata index was created won't be filterable on that property.
51
52	### Index Configuration (Cannot Be Changed Later)
53
54	```bash
55	# Dimensions MUST match your embedding model output:
56	# - Workers AI @cf/baai/bge-base-en-v1.5: 768 dimensions
57	# - OpenAI text-embedding-3-small: 1536 dimensions
58	# - OpenAI text-embedding-3-large: 3072 dimensions
59
60	# Metrics determine similarity calculation:
61	# - cosine: Best for normalized embeddings (most common)
62	# - euclidean: Absolute distance between vectors
63	# - dot-product: For non-normalized vectors
64	```
65
66	## Wrangler Configuration
67
68	wrangler.jsonc:
69	```jsonc
70	{
71	"name": "my-vectorize-worker",
72	"main": "src/index.ts",
73	"compatibility_date": "2025-10-21",
74	"vectorize": [
75	{
76	"binding": "VECTORIZE_INDEX",
77	"index_name": "my-index"
78	}
79	],
80	"ai": {
81	"binding": "AI"
82	}
83	}
84	```
85
86	## TypeScript Types
87
88	```typescript
89	export interface Env {
90	VECTORIZE_INDEX: VectorizeIndex;
91	AI: Ai;
92	}
93
94	interface VectorizeVector {
95	id: string;
96	values: number[] \| Float32Array \| Float64Array;
97	namespace?: string;
98	metadata?: Record<string, string \| number \| boolean \| string[]>;
99	}
100
101	interface VectorizeMatches {
102	matches: Array<{
103	id: string;
104	score: number;
105	values?: number[];
106	metadata?: Record<string, any>;
107	namespace?: string;
108	}>;
109	count: number;
110	}
111	```
112
113	## Common Operations
114
115	### 1. Insert vs Upsert
116
117	```typescript
118	// INSERT: Keeps first insertion if ID exists
119	await env.VECTORIZE_INDEX.insert([
120	{
121	id: "doc-1",
122	values: [0.1, 0.2, 0.3, ...],
123	metadata: { title: "First version" }
124	}
125	]);
126
127	// UPSERT: Overwrites with latest if ID exi

People who liked this also viewed

6 picks

Type

cloudflare-vectorize

15stars7forks

View on GitHub

How to install cloudflare-vectorize?

jackspace/claudeskillz/cloudflare-vectorize

$npx -y skills add jackspace/ClaudeSkillz --skill cloudflare-vectorize

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Cloudflare Vectorize
2
3	Complete implementation guide for Cloudflare Vectorize - a globally distributed vector database for building semantic search, RAG (Retrieval Augmented Generation), and AI-powered applications with Cloudflare Workers.
4
5	Status: Production Ready ✅
6	Last Updated: 2025-10-21
7	Dependencies: cloudflare-worker-base (for Worker setup), cloudflare-workers-ai (for embeddings)
8	Latest Versions: wrangler@4.43.0, @cloudflare/workers-types@4.20251014.0
9	Token Savings: ~65%
10	Errors Prevented: 8
11	Dev Time Saved: ~3 hours
12
13	## What This Skill Provides
14
15	### Core Capabilities
16	- ✅ Index Management: Create, configure, and manage vector indexes
17	- ✅ Vector Operations: Insert, upsert, query, delete, and list vectors
18	- ✅ Metadata Filtering: Advanced filtering with 10 metadata indexes per index
19	- ✅ Semantic Search: Find similar vectors using cosine, euclidean, or dot-product metrics
20	- ✅ RAG Patterns: Complete retrieval-augmented generation workflows
21	- ✅ Workers AI Integration: Native embedding generation with @cf/baai/bge-base-en-v1.5
22	- ✅ OpenAI Integration: Support for text-embedding-3-small/large models
23	- ✅ Document Processing: Text chunking and batch ingestion pipelines
24
25	### Templates Included
26	1. basic-search.ts - Simple vector search with Workers AI
27	2. rag-chat.ts - Full RAG chatbot with context retrieval
28	3. document-ingestion.ts - Document chunking and embedding pipeline
29	4. metadata-filtering.ts - Advanced filtering examples
30
31	## Critical Setup Rules
32
33	### ⚠️ MUST DO BEFORE INSERTING VECTORS
34	```bash
35	# 1. Create the index with FIXED dimensions and metric
36	npx wrangler vectorize create my-index \
37	--dimensions=768 \
38	--metric=cosine
39
40	# 2. Create metadata indexes IMMEDIATELY (before inserting vectors!)
41	npx wrangler vectorize create-metadata-index my-index \
42	--property-name=category \
43	--type=string
44
45	npx wrangler vectorize create-metadata-index my-index \
46	--property-name=timestamp \
47	--type=number
48	```
49
50	Why: Metadata indexes MUST exist before vectors are inserted. Vectors added before a metadata index was created won't be filterable on that property.
51
52	### Index Configuration (Cannot Be Changed Later)
53
54	```bash
55	# Dimensions MUST match your embedding model output:
56	# - Workers AI @cf/baai/bge-base-en-v1.5: 768 dimensions
57	# - OpenAI text-embedding-3-small: 1536 dimensions
58	# - OpenAI text-embedding-3-large: 3072 dimensions
59
60	# Metrics determine similarity calculation:
61	# - cosine: Best for normalized embeddings (most common)
62	# - euclidean: Absolute distance between vectors
63	# - dot-product: For non-normalized vectors
64	```
65
66	## Wrangler Configuration
67
68	wrangler.jsonc:
69	```jsonc
70	{
71	"name": "my-vectorize-worker",
72	"main": "src/index.ts",
73	"compatibility_date": "2025-10-21",
74	"vectorize": [
75	{
76	"binding": "VECTORIZE_INDEX",
77	"index_name": "my-index"
78	}
79	],
80	"ai": {
81	"binding": "AI"
82	}
83	}
84	```
85
86	## TypeScript Types
87
88	```typescript
89	export interface Env {
90	VECTORIZE_INDEX: VectorizeIndex;
91	AI: Ai;
92	}
93
94	interface VectorizeVector {
95	id: string;
96	values: number[] \| Float32Array \| Float64Array;
97	namespace?: string;
98	metadata?: Record<string, string \| number \| boolean \| string[]>;
99	}
100
101	interface VectorizeMatches {
102	matches: Array<{
103	id: string;
104	score: number;
105	values?: number[];
106	metadata?: Record<string, any>;
107	namespace?: string;
108	}>;
109	count: number;
110	}
111	```
112
113	## Common Operations
114
115	### 1. Insert vs Upsert
116
117	```typescript
118	// INSERT: Keeps first insertion if ID exists
119	await env.VECTORIZE_INDEX.insert([
120	{
121	id: "doc-1",
122	values: [0.1, 0.2, 0.3, ...],
123	metadata: { title: "First version" }
124	}
125	]);
126
127	// UPSERT: Overwrites with latest if ID exi

People who liked this also viewed

6 picks

Type