@mastra/brightdata 0.3.1-alpha.0 → 0.3.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +18 -82
  2. package/package.json +4 -4
package/README.md CHANGED
@@ -3,104 +3,40 @@
3
3
  [Bright Data](https://brightdata.com) web search and web fetch tools for [Mastra](https://mastra.ai) agents.
4
4
 
5
5
  Backed by the official [`@brightdata/sdk`](https://github.com/brightdata/sdk-js). Bright Data's SERP API and Web Unlocker bypass bot detection and CAPTCHAs, so the tools work on sites that block typical scrapers.
6
-
7
6
  ## Installation
8
7
 
9
8
  ```bash
10
- npm install @mastra/brightdata zod
9
+ npm install @mastra/brightdata
11
10
  ```
12
11
 
13
- ## Quick Start
12
+ ## Usage
13
+
14
14
 
15
- Use `createBrightDataTools()` to get both tools with a shared configuration:
15
+ Set `BRIGHTDATA_API_TOKEN` and configure the Bright Data zones used by the tools.
16
16
 
17
17
  ```typescript
18
- import { Agent } from '@mastra/core/agent';
19
18
  import { createBrightDataTools } from '@mastra/brightdata';
19
+ import { Agent } from '@mastra/core/agent';
20
20
 
21
- const tools = createBrightDataTools();
22
- // Or pass an explicit API token:
23
- // const tools = createBrightDataTools({ apiKey: 'brd_...' });
21
+ const { webSearch, webFetch } = createBrightDataTools();
24
22
 
25
- const agent = new Agent({
26
- id: 'realtime-information-agent',
27
- name: 'Realtime Information Agent',
28
- instructions:
29
- 'You are a realtime information agent. Use brightdata-search to find pages, and brightdata-fetch to read them.',
30
- model: 'anthropic/claude-sonnet-4-6',
31
- tools,
23
+ export const researchAgent = new Agent({
24
+ id: 'research-agent',
25
+ name: 'Research Agent',
26
+ model: 'openai/gpt-5.6-sol',
27
+ instructions: 'Search for relevant pages, then fetch the best sources before answering.',
28
+ tools: { webSearch, webFetch },
32
29
  });
33
30
  ```
34
31
 
35
- By default the tools read `BRIGHTDATA_API_TOKEN` from your environment. You can also pass `{ apiKey }` explicitly.
36
-
37
- ## Individual Tools
38
-
39
- Each tool can be created independently:
40
-
41
- ```typescript
42
- import { createBrightDataSearchTool, createBrightDataFetchTool } from '@mastra/brightdata';
43
-
44
- const search = createBrightDataSearchTool({ apiKey: 'brd_...' });
45
- const fetch = createBrightDataFetchTool(); // uses BRIGHTDATA_API_TOKEN env var
46
- ```
47
-
48
- ### Web Search (`brightdata-search`)
49
-
50
- ```typescript
51
- import { createBrightDataSearchTool } from '@mastra/brightdata';
52
-
53
- const searchTool = createBrightDataSearchTool();
32
+ ## Documentation
54
33
 
55
- // When called by an agent, accepts:
56
- // - query (required)
57
- // - country: 2-letter code (e.g., 'us', 'gb')
58
- // - start: result offset for pagination (e.g. 10 for the second page of 10 results)
59
- //
60
- // Returns:
61
- // {
62
- // query: string,
63
- // results: Array<{ link, title, description }>,
64
- // currentPage: number
65
- // }
66
- ```
34
+ - [@mastra/brightdata documentation](https://mastra.ai/integrations/tools/brightdata)
67
35
 
68
- ### Web Fetch (`brightdata-fetch`)
36
+ ## Changelog
69
37
 
70
- ```typescript
71
- import { createBrightDataFetchTool } from '@mastra/brightdata';
72
-
73
- const fetchTool = createBrightDataFetchTool();
74
-
75
- // Accepts: url (required)
76
- // Returns: { url, content } // content is Markdown
77
- ```
78
-
79
- ## Configuration
80
-
81
- | Option | Type | Default | Description |
82
- |---|---|---|---|
83
- | `apiKey` | `string` | `process.env.BRIGHTDATA_API_TOKEN` | Your Bright Data API token |
84
-
85
- All tools accept the full `BrightDataClientOptions` from `@brightdata/sdk` (including `timeout`, `webUnlockerZone`, `serpZone`, `rateLimit`, etc.). If no API token is found, the tool throws a clear error at execution time.
86
-
87
- ## RAG Pairing Example
88
-
89
- Combine search and fetch for retrieval-augmented generation:
90
-
91
- ```typescript
92
- import { Agent } from '@mastra/core/agent';
93
- import { createBrightDataTools } from '@mastra/brightdata';
94
-
95
- const agent = new Agent({
96
- id: 'rag-agent',
97
- name: 'Research Assistant',
98
- model: 'anthropic/claude-sonnet-4-6',
99
- instructions: `You are a research assistant. Use brightdata-search to find relevant pages, then use brightdata-fetch to get full Markdown content from the best results.`,
100
- tools: createBrightDataTools(),
101
- });
102
- ```
38
+ See the [package changelog](https://github.com/mastra-ai/mastra/blob/main/integrations/brightdata/CHANGELOG.md) for version history and release notes.
103
39
 
104
- ## License
40
+ ## Support
105
41
 
106
- Apache-2.0
42
+ We have an [open community Discord](https://discord.gg/mastra-ai). Come and say hello and let us know if you have any questions or need any help getting things running.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@mastra/brightdata",
3
- "version": "0.3.1-alpha.0",
3
+ "version": "0.3.1",
4
4
  "description": "Bright Data web search and web fetch tools for Mastra agents",
5
5
  "type": "module",
6
6
  "main": "./dist/index.js",
@@ -51,9 +51,9 @@
51
51
  "typescript": "^7.0.2",
52
52
  "vitest": "4.1.10",
53
53
  "zod": "^4.4.3",
54
- "@internal/lint": "0.0.129",
55
- "@internal/types-builder": "0.0.104",
56
- "@mastra/core": "1.64.0-alpha.2"
54
+ "@internal/lint": "0.0.130",
55
+ "@mastra/core": "1.64.0",
56
+ "@internal/types-builder": "0.0.105"
57
57
  },
58
58
  "scripts": {
59
59
  "build:lib": "tsdown --silent --config tsdown.config.ts",