wuying-agentbay-sdk 0.12.0 → 0.13.1

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 (52) hide show
  1. package/dist/{chunk-BVWUCG4J.mjs → chunk-P2CXYF4T.mjs} +400 -163
  2. package/dist/chunk-P2CXYF4T.mjs.map +1 -0
  3. package/dist/{chunk-SL5GCAQE.cjs → chunk-WVWGLZDT.cjs} +337 -100
  4. package/dist/chunk-WVWGLZDT.cjs.map +1 -0
  5. package/dist/index.cjs +6854 -782
  6. package/dist/index.cjs.map +1 -1
  7. package/dist/index.d.mts +644 -166
  8. package/dist/index.d.ts +644 -166
  9. package/dist/index.mjs +6723 -651
  10. package/dist/index.mjs.map +1 -1
  11. package/dist/{model-LGWQJWKQ.mjs → model-BRLR6F3P.mjs} +16 -2
  12. package/dist/model-KJHN3WYY.cjs +214 -0
  13. package/dist/{model-CNCGFWJH.cjs.map → model-KJHN3WYY.cjs.map} +1 -1
  14. package/docs/api/README.md +6 -0
  15. package/docs/api/browser-use/browser-agent.md +188 -0
  16. package/docs/api/browser-use/browser.md +1 -1
  17. package/docs/api/browser-use/fingerprint.md +154 -0
  18. package/docs/api/codespace/code.md +3 -0
  19. package/docs/api/common-features/advanced/agent.md +7 -63
  20. package/docs/api/common-features/advanced/browser-use-agent.md +118 -0
  21. package/docs/api/common-features/advanced/computer-use-agent.md +85 -0
  22. package/docs/api/common-features/basics/agentbay.md +3 -2
  23. package/docs/api/common-features/basics/command.md +35 -18
  24. package/docs/api/common-features/basics/filesystem.md +36 -0
  25. package/docs/api/common-features/basics/session-params.md +382 -0
  26. package/docs/api/common-features/basics/session.md +0 -2
  27. package/docs/api/computer-use/computer.md +25 -25
  28. package/docs/api/mobile-use/mobile-simulate.md +135 -0
  29. package/docs/examples/browser-use/browser/basic-usage.ts +31 -24
  30. package/docs/examples/browser-use/browser/browser-type-example.ts +3 -4
  31. package/docs/examples/browser-use/browser/captcha_tongcheng.ts +60 -28
  32. package/docs/examples/browser-use/browser/run-2048.ts +47 -37
  33. package/docs/examples/browser-use/browser/run-sudoku.ts +55 -36
  34. package/docs/examples/browser-use/browser/screenshot-example.ts +6 -6
  35. package/docs/examples/browser-use/extension-example/extension-example.ts +1 -2
  36. package/docs/examples/codespace/enhanced_code/index.ts +86 -0
  37. package/docs/examples/common-features/advanced/agent-module-example.ts +1 -1
  38. package/docs/examples/common-features/advanced/archive-upload-mode-example/README.md +1 -1
  39. package/docs/examples/common-features/advanced/archive-upload-mode-example/archive-upload-mode-example.ts +5 -6
  40. package/docs/examples/common-features/basics/archive-upload-mode-example/README.md +1 -1
  41. package/docs/examples/common-features/basics/archive-upload-mode-example/main.ts +1 -1
  42. package/docs/examples/common-features/basics/filesystem-example/filesystem-example.ts +13 -0
  43. package/docs/examples/common-features/basics/filesystem-example/filesystem-filetransfer-example.ts +6 -7
  44. package/docs/examples/common-features/basics/filesystem-example/watch-directory-example.ts +1 -1
  45. package/docs/examples/mobile-use/mobile-get-adb-url/index.ts +1 -1
  46. package/package.json +4 -4
  47. package/dist/chunk-BVWUCG4J.mjs.map +0 -1
  48. package/dist/chunk-SL5GCAQE.cjs.map +0 -1
  49. package/dist/model-CNCGFWJH.cjs +0 -200
  50. package/docs/examples/mobile-use/mobile-get-adb-url/package-lock.json +0 -279
  51. package/docs/examples/mobile-use/mobile-get-adb-url/package.json +0 -18
  52. /package/dist/{model-LGWQJWKQ.mjs.map → model-BRLR6F3P.mjs.map} +0 -0
@@ -5,77 +5,21 @@
5
5
  - [Agent Modules Guide](../../../../../docs/guides/common-features/advanced/agent-modules.md) - Learn about agent modules and custom agents
6
6
 
7
7
  An Agent to manipulate applications to complete specific tasks.
8
+ According to the use scenary, The agent can a browser use agent which is
9
+ specialized for browser automation tasks, The agent also can be a computer
10
+ use agent which is specialized for multiple applications automation tasks.
8
11
 
9
12
  ## Table of contents
10
13
 
11
14
 
12
- ### Methods
15
+ ### Properties
13
16
 
14
- - [executeTask](#executetask)
15
- - [terminateTask](#terminatetask)
16
17
 
17
- ## Methods
18
-
19
- ### executeTask
20
-
21
- ▸ **executeTask**(`task`, `maxTryTimes`): `Promise`\<``ExecutionResult``\>
22
-
23
- Execute a specific task described in human language.
24
-
25
- #### Parameters
26
-
27
- | Name | Type | Description |
28
- | :------ | :------ | :------ |
29
- | `task` | `string` | Task description in human language. |
30
- | `maxTryTimes` | `number` | Maximum number of retry attempts. |
31
-
32
- #### Returns
33
-
34
- `Promise`\<``ExecutionResult``\>
35
-
36
- ExecutionResult containing success status, task output, and error message if any.
37
-
38
- **`Example`**
39
-
40
- ```typescript
41
- const agentBay = new AgentBay({ apiKey: 'your_api_key' });
42
- const result = await agentBay.create({ imageId: 'windows_latest' });
43
- if (result.success) {
44
- const taskResult = await result.session.agent.executeTask('Open notepad', 10);
45
- console.log(`Task status: ${taskResult.taskStatus}`);
46
- await result.session.delete();
47
- }
48
- ```
49
-
50
- ### terminateTask
51
-
52
- ▸ **terminateTask**(`taskId`): `Promise`\<``ExecutionResult``\>
53
-
54
- Terminate a task with a specified task ID.
55
-
56
- #### Parameters
57
-
58
- | Name | Type | Description |
59
- | :------ | :------ | :------ |
60
- | `taskId` | `string` | The ID of the running task. |
61
-
62
- #### Returns
63
-
64
- `Promise`\<``ExecutionResult``\>
65
-
66
- ExecutionResult containing success status, task output, and error message if any.
67
-
68
- **`Example`**
18
+ ## Properties
69
19
 
70
20
  ```typescript
71
- const agentBay = new AgentBay({ apiKey: 'your_api_key' });
72
- const result = await agentBay.create({ imageId: 'windows_latest' });
73
- if (result.success) {
74
- const taskResult = await result.session.agent.executeTask('Open notepad', 5);
75
- const terminateResult = await result.session.agent.terminateTask(taskResult.taskId);
76
- console.log(`Terminated: ${terminateResult.taskStatus}`);
77
- await result.session.delete();
78
- }
21
+ browser: [`BrowserUseAgent`](browser-use-agent.md)
22
+ computer: [`ComputerUseAgent`](computer-use-agent.md)
79
23
  ```
80
24
 
81
25
  ## Related Resources
@@ -0,0 +1,118 @@
1
+ # Class: BrowserUseAgent
2
+
3
+ ## Table of contents
4
+
5
+
6
+ ### Methods
7
+
8
+ - [executeTask](#executetask)
9
+ - [initialize](#initialize)
10
+ - [terminateTask](#terminatetask)
11
+
12
+ ## Methods
13
+
14
+ ### executeTask
15
+
16
+ ▸ **executeTask**(`task`, `maxTryTimes`): `Promise`\<``ExecutionResult``\>
17
+
18
+ Execute a specific task described in human language.
19
+
20
+ #### Parameters
21
+
22
+ | Name | Type | Description |
23
+ | :------ | :------ | :------ |
24
+ | `task` | `string` | Task description in human language. |
25
+ | `maxTryTimes` | `number` | Maximum number of retry attempts. |
26
+
27
+ #### Returns
28
+
29
+ `Promise`\<``ExecutionResult``\>
30
+
31
+ ExecutionResult containing success status, task output, and
32
+ error message if any.
33
+
34
+ **`Example`**
35
+
36
+ ```typescript
37
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
38
+ const result = await agentBay.create({ imageId: 'linux_latest' });
39
+ if (result.success) {
40
+ const taskResult = await result.session.agent.browser.executeTask(
41
+ 'Navigate to baidu and query the weather of Shanghai',
42
+ 10
43
+ );
44
+ console.log(`Task status: ${taskResult.taskStatus}`);
45
+ await result.session.delete();
46
+ }
47
+ ```
48
+
49
+ ### initialize
50
+
51
+ ▸ **initialize**(`options`): `Promise`\<``InitializationResult``\>
52
+
53
+ Initialize the browser agent with specific options.
54
+
55
+ #### Parameters
56
+
57
+ | Name | Type | Description |
58
+ | :------ | :------ | :------ |
59
+ | `options` | ``AgentOptions`` | agent initialization options |
60
+
61
+ #### Returns
62
+
63
+ `Promise`\<``InitializationResult``\>
64
+
65
+ InitializationResult containing success status, task output,
66
+ and error message if any.
67
+
68
+ **`Example`**
69
+
70
+ ```typescript
71
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
72
+ const result = await agentBay.create({ imageId: 'linux_latest' });
73
+ if (result.success) {
74
+ options:AgentOptions = new AgentOptions(use_vision=False,
75
+ output_schema=""); const initResult = await
76
+ result.session.agent.browser.initialize(options); console.log(`Initialize
77
+ success: ${initResult.success}`); await result.session.delete();
78
+ }
79
+ ```
80
+
81
+ ___
82
+
83
+ ### terminateTask
84
+
85
+ ▸ **terminateTask**(`taskId`): `Promise`\<``ExecutionResult``\>
86
+
87
+ Terminate a task with a specified task ID.
88
+
89
+ #### Parameters
90
+
91
+ | Name | Type | Description |
92
+ | :------ | :------ | :------ |
93
+ | `taskId` | `string` | The ID of the running task. |
94
+
95
+ #### Returns
96
+
97
+ `Promise`\<``ExecutionResult``\>
98
+
99
+ ExecutionResult containing success status, task output, and
100
+ error message if any.
101
+
102
+ **`Example`**
103
+
104
+ ```typescript
105
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
106
+ const result = await agentBay.create({ imageId: 'linux_latest' });
107
+ if (result.success) {
108
+ const taskResult = await result.session.agent.browser.executeTask(
109
+ 'Navigate to baidu and query the weather of Shanghai',
110
+ 10
111
+ );
112
+ const terminateResult = await result.session.agent.browser.terminateTask(
113
+ taskResult.taskId
114
+ );
115
+ console.log(`Terminated: ${terminateResult.taskStatus}`);
116
+ await result.session.delete();
117
+ }
118
+ ```
@@ -0,0 +1,85 @@
1
+ # Class: ComputerUseAgent
2
+
3
+ An Agent to perform tasks on the computer.
4
+
5
+ ## Table of contents
6
+
7
+
8
+ ### Methods
9
+
10
+ - [executeTask](#executetask)
11
+ - [terminateTask](#terminatetask)
12
+
13
+ ## Methods
14
+
15
+ ### executeTask
16
+
17
+ ▸ **executeTask**(`task`, `maxTryTimes`): `Promise`\<``ExecutionResult``\>
18
+
19
+ Execute a specific task described in human language.
20
+
21
+ #### Parameters
22
+
23
+ | Name | Type | Description |
24
+ | :------ | :------ | :------ |
25
+ | `task` | `string` | Task description in human language. |
26
+ | `maxTryTimes` | `number` | Maximum number of retry attempts. |
27
+
28
+ #### Returns
29
+
30
+ `Promise`\<``ExecutionResult``\>
31
+
32
+ ExecutionResult containing success status, task output, and error
33
+ message if any.
34
+
35
+ **`Example`**
36
+
37
+ ```typescript
38
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
39
+ const result = await agentBay.create({ imageId: 'windows_latest' });
40
+ if (result.success) {
41
+ const taskResult = await result.session.agent.computer.executeTask(
42
+ 'Open notepad',
43
+ 10
44
+ );
45
+ console.log(`Task status: ${taskResult.taskStatus}`);
46
+ await result.session.delete();
47
+ }
48
+ ```
49
+
50
+ ### terminateTask
51
+
52
+ ▸ **terminateTask**(`taskId`): `Promise`\<``ExecutionResult``\>
53
+
54
+ Terminate a task with a specified task ID.
55
+
56
+ #### Parameters
57
+
58
+ | Name | Type | Description |
59
+ | :------ | :------ | :------ |
60
+ | `taskId` | `string` | The ID of the running task. |
61
+
62
+ #### Returns
63
+
64
+ `Promise`\<``ExecutionResult``\>
65
+
66
+ ExecutionResult containing success status, task output, and
67
+ error message if any.
68
+
69
+ **`Example`**
70
+
71
+ ```typescript
72
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
73
+ const result = await agentBay.create({ imageId: 'windows_latest' });
74
+ if (result.success) {
75
+ const taskResult = await result.session.agent.computer.executeTask(
76
+ 'Open notepad',
77
+ 5
78
+ );
79
+ const terminateResult = await result.session.agent.computer.terminateTask(
80
+ taskResult.taskId
81
+ );
82
+ console.log(`Terminated: ${terminateResult.taskStatus}`);
83
+ await result.session.delete();
84
+ }
85
+ ```
@@ -24,6 +24,7 @@ Main class for interacting with the AgentBay cloud runtime environment.
24
24
  ## Properties
25
25
 
26
26
  ```typescript
27
+ client: ``Client``
27
28
  context: [`ContextService`](context.md)
28
29
  ```
29
30
 
@@ -32,7 +33,7 @@ context: [`ContextService`](context.md)
32
33
 
33
34
  ### create
34
35
 
35
- ▸ **create**(`params?`): `Promise`\<`SessionResult`\>
36
+ ▸ **create**(`params`): `Promise`\<`SessionResult`\>
36
37
 
37
38
  Creates a new AgentBay session with specified configuration.
38
39
 
@@ -40,7 +41,7 @@ Creates a new AgentBay session with specified configuration.
40
41
 
41
42
  | Name | Type | Description |
42
43
  | :------ | :------ | :------ |
43
- | `params` | ``CreateSessionParams`` | Configuration parameters for the session: - labels: Key-value pairs for session metadata - imageId: Custom image ID for the session environment - contextSync: Array of context synchronization configurations - browserContext: Browser-specific context configuration - isVpc: Whether to create a VPC session - policyId: Security policy ID - enableBrowserReplay: Enable browser session recording - extraConfigs: Additional configuration options - framework: Framework identifier for tracking |
44
+ | `params` | [`CreateSessionParams`](session-params.md) \| `CreateSeesionWithParams` | Configuration parameters for the session: - labels: Key-value pairs for session metadata - imageId: Custom image ID for the session environment - contextSync: Array of context synchronization configurations - browserContext: Browser-specific context configuration - isVpc: Whether to create a VPC session - policyId: Security policy ID - enableBrowserReplay: Enable browser session recording - extraConfigs: Additional configuration options - framework: Framework identifier for tracking |
44
45
 
45
46
  #### Returns
46
47
 
@@ -7,7 +7,7 @@
7
7
  ## Overview
8
8
 
9
9
  The Command module provides methods for executing shell commands within a session in the AgentBay cloud environment.
10
- It supports both synchronous command execution with configurable timeouts.
10
+ Commands support configurable timeouts and optional working directory or environment settings.
11
11
 
12
12
  Handles command execution operations in the AgentBay cloud environment.
13
13
 
@@ -22,24 +22,34 @@ Handles command execution operations in the AgentBay cloud environment.
22
22
 
23
23
  ### executeCommand
24
24
 
25
- ▸ **executeCommand**(`command`, `timeoutMs?`): `Promise`\<`CommandResult`\>
25
+ ▸ **executeCommand**(`command`, `timeoutMs?`, `cwd?`, `envs?`): `Promise`\<`CommandResult`\>
26
26
 
27
- Executes a shell command in the session environment.
27
+ Execute a shell command with optional working directory and environment variables.
28
+
29
+ Executes a shell command in the session environment with configurable timeout,
30
+ working directory, and environment variables. The command runs with session
31
+ user permissions in a Linux shell environment.
28
32
 
29
33
  #### Parameters
30
34
 
31
35
  | Name | Type | Default value | Description |
32
36
  | :------ | :------ | :------ | :------ |
33
- | `command` | `string` | `undefined` | The shell command to execute. |
34
- | `timeoutMs` | `number` | `1000` | Timeout in milliseconds. Defaults to 1000ms. |
37
+ | `command` | `string` | `undefined` | The shell command to execute |
38
+ | `timeoutMs` | `number` | `1000` | Timeout in milliseconds (default: 1000ms/1s). Maximum allowed timeout is 50000ms (50s). If a larger value is provided, it will be automatically limited to 50000ms |
39
+ | `cwd?` | `string` | `undefined` | The working directory for command execution. If not specified, the command runs in the default session directory |
40
+ | `envs?` | `Record`\<`string`, `string`\> | `undefined` | Environment variables as a dictionary of key-value pairs. These variables are set for the command execution only |
35
41
 
36
42
  #### Returns
37
43
 
38
44
  `Promise`\<`CommandResult`\>
39
45
 
40
46
  Promise resolving to CommandResult containing:
41
- - success: Whether the command executed successfully
42
- - output: Combined stdout and stderr output
47
+ - success: Whether the command executed successfully (exitCode === 0)
48
+ - output: Command output for backward compatibility (stdout + stderr)
49
+ - exitCode: The exit code of the command execution (0 for success)
50
+ - stdout: Standard output from the command execution
51
+ - stderr: Standard error from the command execution
52
+ - traceId: Trace ID for error tracking (only present when exitCode !== 0)
43
53
  - requestId: Unique identifier for this API request
44
54
  - errorMessage: Error description if execution failed
45
55
 
@@ -49,23 +59,30 @@ Promise resolving to CommandResult containing:
49
59
  const agentBay = new AgentBay({ apiKey: 'your_api_key' });
50
60
  const result = await agentBay.create();
51
61
  if (result.success) {
52
- const cmdResult = await result.session.command.executeCommand('echo "Hello"', 3000);
62
+ const cmdResult = await result.session.command.executeCommand('echo "Hello"', 5000);
53
63
  console.log('Command output:', cmdResult.output);
64
+ console.log('Exit code:', cmdResult.exitCode);
65
+ console.log('Stdout:', cmdResult.stdout);
54
66
  await result.session.delete();
55
67
  }
56
68
  ```
57
69
 
58
- **`Remarks`**
59
-
60
- **Behavior:**
61
- - Executes in a Linux shell environment
62
- - Combines stdout and stderr in the output
63
- - Default timeout is 1000ms (1 second)
64
- - Command runs with session user permissions
65
-
66
- **`See`**
70
+ **`Example`**
67
71
 
68
- [FileSystem.readFile](filesystem.md#readfile), [FileSystem.writeFile](filesystem.md#writefile)
72
+ ```typescript
73
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
74
+ const result = await agentBay.create();
75
+ if (result.success) {
76
+ const cmdResult = await result.session.command.executeCommand(
77
+ 'pwd',
78
+ 5000,
79
+ '/tmp',
80
+ { TEST_VAR: 'test_value' }
81
+ );
82
+ console.log('Working directory:', cmdResult.stdout);
83
+ await result.session.delete();
84
+ }
85
+ ```
69
86
 
70
87
  ## Best Practices
71
88
 
@@ -12,6 +12,7 @@ Handles file operations in the AgentBay cloud environment.
12
12
  ### Methods
13
13
 
14
14
  - [createDirectory](#createdirectory)
15
+ - [deleteFile](#deletefile)
15
16
  - [downloadFile](#downloadfile)
16
17
  - [editFile](#editfile)
17
18
  - [listDirectory](#listdirectory)
@@ -58,6 +59,41 @@ if (result.success) {
58
59
 
59
60
  ___
60
61
 
62
+ ### deleteFile
63
+
64
+ ▸ **deleteFile**(`path`): `Promise`\<`BoolResult`\>
65
+
66
+ Deletes a file at the specified path.
67
+ Corresponds to Python's delete_file() method
68
+
69
+ #### Parameters
70
+
71
+ | Name | Type | Description |
72
+ | :------ | :------ | :------ |
73
+ | `path` | `string` | Path to the file to delete. |
74
+
75
+ #### Returns
76
+
77
+ `Promise`\<`BoolResult`\>
78
+
79
+ BoolResult with deletion result and requestId
80
+
81
+ **`Example`**
82
+
83
+ ```typescript
84
+ const agentBay = new AgentBay({ apiKey: 'your_api_key' });
85
+ const result = await agentBay.create();
86
+ if (result.success) {
87
+ const session = result.session;
88
+ await session.fileSystem.writeFile('/tmp/to_delete.txt', 'hello');
89
+ const deleteResult = await session.fileSystem.deleteFile('/tmp/to_delete.txt');
90
+ console.log('File deleted:', deleteResult.success);
91
+ await session.delete();
92
+ }
93
+ ```
94
+
95
+ ___
96
+
61
97
  ### downloadFile
62
98
 
63
99
  ▸ **downloadFile**(`remotePath`, `localPath`, `options?`): `Promise`\<`any`\>