cognitrix 0.2.0__tar.gz → 0.2.1__tar.gz
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.
- {cognitrix-0.2.0 → cognitrix-0.2.1}/PKG-INFO +15 -13
- cognitrix-0.2.1/README.md +77 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/agents/base.py +3 -3
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/agents/templates.py +2 -15
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/app.py +0 -6
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/cli.py +9 -3
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/config.py +1 -1
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tools/__init__.py +3 -1
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tools/misc.py +105 -11
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tools/tool.py +7 -6
- {cognitrix-0.2.0 → cognitrix-0.2.1}/pyproject.toml +1 -1
- cognitrix-0.2.0/README.md +0 -75
- {cognitrix-0.2.0 → cognitrix-0.2.1}/LICENSE +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/__init__.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/agents/__init__.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/agents/assistant.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/agents/core.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/__init__.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/anthropic_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/base.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/clarifai_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/cohere_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/google_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/groq_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/openai_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/llms/together_llm.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tasks/__init__.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tasks/base.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tools/base.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tools/python.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/tools/serpapi.py +0 -0
- {cognitrix-0.2.0 → cognitrix-0.2.1}/cognitrix/utils/__init__.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.1
|
|
2
2
|
Name: cognitrix
|
|
3
|
-
Version: 0.2.
|
|
3
|
+
Version: 0.2.1
|
|
4
4
|
Summary: Package for creating AI Agents using llms
|
|
5
5
|
Home-page: https://github.com/theonlyamos/cognitrix
|
|
6
6
|
License: Apache2
|
|
@@ -47,19 +47,21 @@ Description-Content-Type: text/markdown
|
|
|
47
47
|
Cognitrix is an open-source autonous AI agents orchestrator built in Python. It allows you to create and manage AI agents easily.
|
|
48
48
|
|
|
49
49
|
### Features
|
|
50
|
-
- Create and
|
|
51
|
-
-
|
|
52
|
-
- Modular
|
|
53
|
-
-
|
|
54
|
-
-
|
|
50
|
+
- **`Agent Creation and Management`:** Create, list, and load AI agents with customizable names, tasks, and configurations.
|
|
51
|
+
- **`LLM Integration`:** Seamlessly integrate with multiple LLM providers, including Anthropic (Claude), Cohere, Groq, Google, OpenAI, and Together.
|
|
52
|
+
- **`Modular Architecture`:** Easily extend the framework by adding new tools, agents, and LLM integrations.
|
|
53
|
+
- **`Conversational Interface`:** Interact with AI agents through a command-line interface, providing queries and receiving responses.
|
|
54
|
+
- **`Tool Integration`:** Agents can utilize a variety of tools, such as calculators, web searches, file system browsers, and more.
|
|
55
|
+
- **`Autonomous Agent Mode`:** Enable agents to operate autonomously, visually perceiving the screen, interacting with UI elements, and performing tasks.
|
|
56
|
+
- **`Multimodal Support`:** Agents can handle both text and image inputs/outputs, enabling multimodal interactions.
|
|
55
57
|
|
|
56
58
|
### Architecture
|
|
57
|
-
|
|
59
|
+
Cognitrix's architecture is designed to be modular and extensible, with core components including:
|
|
58
60
|
|
|
59
|
-
|
|
60
|
-
|
|
61
|
-
|
|
62
|
-
|
|
61
|
+
1. **`Agents`:** The base Agent class and specialized classes like AIAssistant for creating and managing AI agents.
|
|
62
|
+
2. **`LLMs`:** A collection of classes for integrating with various LLM providers, such as Cohere, OpenAI, Claude, and more.
|
|
63
|
+
3. **`Tools`:** A set of tools that agents can utilize, including calculators, web searches, file system browsers, and more.
|
|
64
|
+
4. **`Templates`:** Customizable prompt templates for guiding the behavior and output formats of LLMs.
|
|
63
65
|
|
|
64
66
|
The architecture is highly modular and extensible. New tools and capabilities can be easily added.
|
|
65
67
|
|
|
@@ -116,5 +118,5 @@ Cognitrix is open source and contributions are welcome. See [CONTRIBUTING.md](CO
|
|
|
116
118
|
### License
|
|
117
119
|
This project is licensed under the MIT license. See [LICENSE.md](LICENSE.md) for more details.
|
|
118
120
|
|
|
119
|
-
###
|
|
120
|
-
Cognitrix was created by [Amos Amissah](https://github.com/theonlyamos).
|
|
121
|
+
### Acknowledgments
|
|
122
|
+
Cognitrix was created by [Amos Amissah](https://github.com/theonlyamos) and is heavily inspired by projects like AutoGPT and EngineerGPT. Special thanks to the open-source community for their contributions and the AI companies providing LLM APIs.
|
|
@@ -0,0 +1,77 @@
|
|
|
1
|
+
### Cognitrix
|
|
2
|
+
Cognitrix is an open-source autonous AI agents orchestrator built in Python. It allows you to create and manage AI agents easily.
|
|
3
|
+
|
|
4
|
+
### Features
|
|
5
|
+
- **`Agent Creation and Management`:** Create, list, and load AI agents with customizable names, tasks, and configurations.
|
|
6
|
+
- **`LLM Integration`:** Seamlessly integrate with multiple LLM providers, including Anthropic (Claude), Cohere, Groq, Google, OpenAI, and Together.
|
|
7
|
+
- **`Modular Architecture`:** Easily extend the framework by adding new tools, agents, and LLM integrations.
|
|
8
|
+
- **`Conversational Interface`:** Interact with AI agents through a command-line interface, providing queries and receiving responses.
|
|
9
|
+
- **`Tool Integration`:** Agents can utilize a variety of tools, such as calculators, web searches, file system browsers, and more.
|
|
10
|
+
- **`Autonomous Agent Mode`:** Enable agents to operate autonomously, visually perceiving the screen, interacting with UI elements, and performing tasks.
|
|
11
|
+
- **`Multimodal Support`:** Agents can handle both text and image inputs/outputs, enabling multimodal interactions.
|
|
12
|
+
|
|
13
|
+
### Architecture
|
|
14
|
+
Cognitrix's architecture is designed to be modular and extensible, with core components including:
|
|
15
|
+
|
|
16
|
+
1. **`Agents`:** The base Agent class and specialized classes like AIAssistant for creating and managing AI agents.
|
|
17
|
+
2. **`LLMs`:** A collection of classes for integrating with various LLM providers, such as Cohere, OpenAI, Claude, and more.
|
|
18
|
+
3. **`Tools`:** A set of tools that agents can utilize, including calculators, web searches, file system browsers, and more.
|
|
19
|
+
4. **`Templates`:** Customizable prompt templates for guiding the behavior and output formats of LLMs.
|
|
20
|
+
|
|
21
|
+
The architecture is highly modular and extensible. New tools and capabilities can be easily added.
|
|
22
|
+
|
|
23
|
+
### Getting Started
|
|
24
|
+
|
|
25
|
+
### Installation
|
|
26
|
+
```bash
|
|
27
|
+
pip install cognitrix
|
|
28
|
+
```
|
|
29
|
+
or
|
|
30
|
+
```bash
|
|
31
|
+
pip install https://github.com/theonlyamos/cognitrix/archive/main.zip
|
|
32
|
+
```
|
|
33
|
+
|
|
34
|
+
### Usage
|
|
35
|
+
Run with default settings
|
|
36
|
+
```bash
|
|
37
|
+
cognitrix
|
|
38
|
+
```
|
|
39
|
+
|
|
40
|
+
List supported platforms
|
|
41
|
+
```bash
|
|
42
|
+
cognitrix --platforms
|
|
43
|
+
```
|
|
44
|
+
|
|
45
|
+
Run with specific platform
|
|
46
|
+
```bash
|
|
47
|
+
cognitrix --platform <platform_name>
|
|
48
|
+
```
|
|
49
|
+
|
|
50
|
+
Create a new agent
|
|
51
|
+
```bash
|
|
52
|
+
cognitrix agents --new
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
List created agents
|
|
56
|
+
```bash
|
|
57
|
+
cognitrix agents
|
|
58
|
+
```
|
|
59
|
+
|
|
60
|
+
Run with specific agent
|
|
61
|
+
```bash
|
|
62
|
+
cognitrix --agent <agent_name>
|
|
63
|
+
```
|
|
64
|
+
|
|
65
|
+
Print help message
|
|
66
|
+
```bash
|
|
67
|
+
cognitrix --help
|
|
68
|
+
```
|
|
69
|
+
|
|
70
|
+
### Contributing
|
|
71
|
+
Cognitrix is open source and contributions are welcome. See [CONTRIBUTING.md](CONTRIBUTING.md) for more details.
|
|
72
|
+
|
|
73
|
+
### License
|
|
74
|
+
This project is licensed under the MIT license. See [LICENSE.md](LICENSE.md) for more details.
|
|
75
|
+
|
|
76
|
+
### Acknowledgments
|
|
77
|
+
Cognitrix was created by [Amos Amissah](https://github.com/theonlyamos) and is heavily inspired by projects like AutoGPT and EngineerGPT. Special thanks to the open-source community for their contributions and the AI companies providing LLM APIs.
|
|
@@ -12,7 +12,7 @@ from pydantic import BaseModel, Field
|
|
|
12
12
|
from ..llms.base import LLM
|
|
13
13
|
from ..tools.base import Tool
|
|
14
14
|
from ..tasks.base import Task
|
|
15
|
-
from ..agents.templates import
|
|
15
|
+
from ..agents.templates import ASSISTANT_TEMPLATE
|
|
16
16
|
from ..config import AGENTS_FILE
|
|
17
17
|
|
|
18
18
|
logging.basicConfig(
|
|
@@ -35,7 +35,7 @@ class Agent(BaseModel):
|
|
|
35
35
|
tools: List[Tool] = Field(default=[])
|
|
36
36
|
"""Tools to be used by agent"""
|
|
37
37
|
|
|
38
|
-
prompt_template: str = Field(default=
|
|
38
|
+
prompt_template: str = Field(default=ASSISTANT_TEMPLATE)
|
|
39
39
|
"""Base system prompt template"""
|
|
40
40
|
|
|
41
41
|
verbose: bool = Field(default=False)
|
|
@@ -206,7 +206,7 @@ class Agent(BaseModel):
|
|
|
206
206
|
Initialize the llm
|
|
207
207
|
"""
|
|
208
208
|
self.format_system_prompt()
|
|
209
|
-
|
|
209
|
+
|
|
210
210
|
query: str|dict = input("\nUser (q to quit): ")
|
|
211
211
|
while query:
|
|
212
212
|
try:
|
|
@@ -74,20 +74,6 @@ Context: The AI assistant has access to the user's computer and the internet. Th
|
|
|
74
74
|
The AI assistant will answer the user's question to the best of its ability, using its knowledge and access to tools provided by the user.
|
|
75
75
|
The AI assistant has access to the following tools.
|
|
76
76
|
|
|
77
|
-
You are an expert at world knowledge.
|
|
78
|
-
Your task is to step back and paraphrase a question to a more generic
|
|
79
|
-
step-back question, which is easier to answer.
|
|
80
|
-
|
|
81
|
-
Here are a few examples:
|
|
82
|
-
Original Question: Which position did Knox Cunningham hold from May 1955 to Apr 1956?
|
|
83
|
-
Stepback Question: Which positions have Knox Cunning- ham held in his career?
|
|
84
|
-
|
|
85
|
-
Original Question: Who was the spouse of Anna Karina from 1968 to 1974?
|
|
86
|
-
Stepback Question: Who were the spouses of Anna Karina?
|
|
87
|
-
|
|
88
|
-
Original Question: Which team did Thierry Audel play for from 2007 to 2008?
|
|
89
|
-
Stepback Question: Which teams did Thierry Audel play for in his career?
|
|
90
|
-
|
|
91
77
|
Remember, functions calls will be processed by the user and the result returned to the AI Asssistant as the next input.
|
|
92
78
|
A function name is a name of a tool available to the AI Assistant.
|
|
93
79
|
Whenever there is a function call, always wait for the answer from the user. Do not try to answer that query yourself.
|
|
@@ -130,7 +116,8 @@ if it's the final anwers or
|
|
|
130
116
|
}
|
|
131
117
|
if the assistant needs to use a tool to answer the user's query.
|
|
132
118
|
Don't forget to present the answer to a function call to the user in an informative manner.
|
|
133
|
-
|
|
119
|
+
You should always break down complex tasks into smaller easier ones and perform them one by one.
|
|
120
|
+
You should always check if a task is complete by taking a screenshot of the screen.
|
|
134
121
|
Begin:
|
|
135
122
|
"""
|
|
136
123
|
|
|
@@ -8,16 +8,10 @@ from tools import (
|
|
|
8
8
|
InternetBrowser, tool
|
|
9
9
|
)
|
|
10
10
|
|
|
11
|
-
@tool
|
|
12
|
-
def print_hello():
|
|
13
|
-
"""Returns hello world"""
|
|
14
|
-
return 'hello world'
|
|
15
|
-
|
|
16
11
|
if __name__ == "__main__":
|
|
17
12
|
llm = Cohere()
|
|
18
13
|
# llm = TogetherLLM()
|
|
19
14
|
assistant = AIAssistant(llm=llm, name='Adam')
|
|
20
|
-
assistant.add_tool(print_hello())
|
|
21
15
|
assistant.add_tool(Calculator())
|
|
22
16
|
assistant.add_tool(YoutubePlayer())
|
|
23
17
|
assistant.add_tool(WorldNews())
|
|
@@ -14,8 +14,9 @@ from .tools import (
|
|
|
14
14
|
Calculator, YoutubePlayer,
|
|
15
15
|
WorldNews, PythonREPL,
|
|
16
16
|
FSBrowser, SearchTool,
|
|
17
|
-
InternetBrowser,
|
|
18
|
-
|
|
17
|
+
InternetBrowser, take_screenshot,
|
|
18
|
+
text_input, key_press, mouse_click,
|
|
19
|
+
mouse_double_click, mouse_right_click
|
|
19
20
|
)
|
|
20
21
|
|
|
21
22
|
from .config import VERSION
|
|
@@ -98,7 +99,12 @@ def start(args: Namespace):
|
|
|
98
99
|
if assistant:
|
|
99
100
|
assistant.llm = platform
|
|
100
101
|
assistant.name = args.name
|
|
101
|
-
assistant.add_tool(take_screenshot
|
|
102
|
+
assistant.add_tool(take_screenshot)
|
|
103
|
+
assistant.add_tool(text_input)
|
|
104
|
+
assistant.add_tool(key_press)
|
|
105
|
+
assistant.add_tool(mouse_click)
|
|
106
|
+
assistant.add_tool(mouse_double_click)
|
|
107
|
+
assistant.add_tool(mouse_right_click)
|
|
102
108
|
# assistant.add_tool(Calculator())
|
|
103
109
|
# assistant.add_tool(YoutubePlayer())
|
|
104
110
|
# assistant.add_tool(WorldNews())
|
|
@@ -3,7 +3,9 @@ from ..tools.tool import tool
|
|
|
3
3
|
from ..tools.misc import (
|
|
4
4
|
Calculator, YoutubePlayer,
|
|
5
5
|
WorldNews, FSBrowser,
|
|
6
|
-
InternetBrowser, take_screenshot
|
|
6
|
+
InternetBrowser, take_screenshot,
|
|
7
|
+
text_input, key_press, mouse_click,
|
|
8
|
+
mouse_double_click, mouse_right_click
|
|
7
9
|
)
|
|
8
10
|
|
|
9
11
|
from ..tools.python import PythonREPL
|
|
@@ -395,20 +395,114 @@ def take_screenshot():
|
|
|
395
395
|
}"""
|
|
396
396
|
screenshot = pyautogui.screenshot()
|
|
397
397
|
|
|
398
|
-
|
|
399
|
-
# width, height = screenshot.size
|
|
400
|
-
# new_width = int(width * 0.5)
|
|
401
|
-
# new_height = int(height * 0.5)
|
|
402
|
-
# screenshot = screenshot.resize((new_width, new_height), Image.Resampling.LANCZOS)
|
|
398
|
+
return ['image', screenshot]
|
|
403
399
|
|
|
404
|
-
|
|
405
|
-
|
|
406
|
-
|
|
400
|
+
@tool
|
|
401
|
+
def text_input(text: str):
|
|
402
|
+
"""Use this tool to take make text inputs.
|
|
403
|
+
Args:
|
|
404
|
+
text (str): Text to input.
|
|
405
|
+
|
|
406
|
+
Returns:
|
|
407
|
+
str: Text input completed.
|
|
408
|
+
|
|
409
|
+
Example:
|
|
410
|
+
User: write hello world
|
|
411
|
+
AI Assistant: {
|
|
412
|
+
"type": "function_call",
|
|
413
|
+
"function": "Text Input",
|
|
414
|
+
"arguments": ["hello world"]
|
|
415
|
+
}
|
|
416
|
+
"""
|
|
417
|
+
screenshot = pyautogui.write(text, 0.25)
|
|
418
|
+
return 'Text input completed'
|
|
407
419
|
|
|
408
|
-
|
|
409
|
-
|
|
420
|
+
@tool
|
|
421
|
+
def key_press(key: str):
|
|
422
|
+
"""Use this tool to take make key presses.
|
|
423
|
+
Args:
|
|
424
|
+
key (str): Name of key to press.
|
|
410
425
|
|
|
411
|
-
|
|
426
|
+
Returns:
|
|
427
|
+
str: Keypress completed.
|
|
428
|
+
|
|
429
|
+
Example:
|
|
430
|
+
User: Press windows key
|
|
431
|
+
AI Assistant: {
|
|
432
|
+
"type": "function_call",
|
|
433
|
+
"function": "Key Press",
|
|
434
|
+
"arguments": ["win"]
|
|
435
|
+
}
|
|
436
|
+
"""
|
|
437
|
+
screenshot = pyautogui.press(key.lower())
|
|
438
|
+
|
|
439
|
+
return 'Keypress completed'
|
|
440
|
+
|
|
441
|
+
@tool
|
|
442
|
+
def mouse_click(x: int, y: int):
|
|
443
|
+
"""Use this tool to take make mouse clicks.
|
|
444
|
+
Args:
|
|
445
|
+
x (int): X coordinate of mouse click.
|
|
446
|
+
y (int): y coordinate of mouse click.
|
|
447
|
+
|
|
448
|
+
Returns:
|
|
449
|
+
str: Mouse click completed.
|
|
450
|
+
|
|
451
|
+
Example:
|
|
452
|
+
User: Click on Brave icon
|
|
453
|
+
AI Assistant: {
|
|
454
|
+
"type": "function_call",
|
|
455
|
+
"function": "Mouse Click",
|
|
456
|
+
"arguments": ["123", "456"]
|
|
457
|
+
}
|
|
458
|
+
"""
|
|
459
|
+
screenshot = pyautogui.click(x, y)
|
|
460
|
+
|
|
461
|
+
return 'Mouse Click completed'
|
|
462
|
+
|
|
463
|
+
@tool
|
|
464
|
+
def mouse_double_click(x: int, y: int):
|
|
465
|
+
"""Use this tool to take make mouse double clicks.
|
|
466
|
+
Args:
|
|
467
|
+
x (int): X coordinate of mouse click.
|
|
468
|
+
y (int): y coordinate of mouse click.
|
|
469
|
+
|
|
470
|
+
Returns:
|
|
471
|
+
str: Mouse double-click completed.
|
|
472
|
+
|
|
473
|
+
Example:
|
|
474
|
+
User: Click on Brave icon
|
|
475
|
+
AI Assistant: {
|
|
476
|
+
"type": "function_call",
|
|
477
|
+
"function": "Mouse Double Click",
|
|
478
|
+
"arguments": ["123", "456"]
|
|
479
|
+
}
|
|
480
|
+
"""
|
|
481
|
+
screenshot = pyautogui.doubleClick(x, y)
|
|
482
|
+
|
|
483
|
+
return 'Mouse double-click completed.'
|
|
484
|
+
|
|
485
|
+
@tool
|
|
486
|
+
def mouse_right_click(x: int, y: int):
|
|
487
|
+
"""Use this tool to take make mouse right clicks.
|
|
488
|
+
Args:
|
|
489
|
+
x (int): X coordinate of mouse click.
|
|
490
|
+
y (int): y coordinate of mouse click.
|
|
491
|
+
|
|
492
|
+
Returns:
|
|
493
|
+
str: Mouse right-click completed.
|
|
494
|
+
|
|
495
|
+
Example:
|
|
496
|
+
User: Right-click on Brave icon
|
|
497
|
+
AI Assistant: {
|
|
498
|
+
"type": "function_call",
|
|
499
|
+
"function": "Mouse Right Click",
|
|
500
|
+
"arguments": ["123", "456"]
|
|
501
|
+
}
|
|
502
|
+
"""
|
|
503
|
+
screenshot = pyautogui.rightClick(x, y)
|
|
504
|
+
|
|
505
|
+
return 'Mouse double-click completed.'
|
|
412
506
|
|
|
413
507
|
# @tool
|
|
414
508
|
# def get_ui_elements_coordinates():
|
|
@@ -9,17 +9,18 @@ def tool(func: Callable, *args, **kwargs):
|
|
|
9
9
|
return func(*args, **kwargs)
|
|
10
10
|
|
|
11
11
|
class GenericTool(Tool):
|
|
12
|
-
|
|
13
|
-
description: str = str(func.__doc__)
|
|
14
|
-
|
|
15
|
-
def run(self):
|
|
12
|
+
def run(self, *args, **kwargs):
|
|
16
13
|
if inspect.iscoroutinefunction(func):
|
|
17
14
|
raise NotImplementedError("Asynchronous execution is not supported for coroutine functions")
|
|
18
15
|
return wrapper(*args, **kwargs)
|
|
19
16
|
|
|
20
|
-
async def arun(self):
|
|
17
|
+
async def arun(self, *args, **kwargs):
|
|
21
18
|
if inspect.iscoroutinefunction(func):
|
|
22
19
|
return await wrapper(*args, **kwargs)
|
|
23
20
|
raise NotImplementedError("Synchronous execution is not supported for synchronous functions")
|
|
24
21
|
|
|
25
|
-
|
|
22
|
+
new_tool = GenericTool(
|
|
23
|
+
name=' '.join(func.__name__.split('_')).title(),
|
|
24
|
+
description=str(func.__doc__)
|
|
25
|
+
)
|
|
26
|
+
return new_tool
|
cognitrix-0.2.0/README.md
DELETED
|
@@ -1,75 +0,0 @@
|
|
|
1
|
-
### Cognitrix
|
|
2
|
-
Cognitrix is an open-source autonous AI agents orchestrator built in Python. It allows you to create and manage AI agents easily.
|
|
3
|
-
|
|
4
|
-
### Features
|
|
5
|
-
- Create and manage AI agents.
|
|
6
|
-
- Conversational interface using chat
|
|
7
|
-
- Modular architecture for easy extensibility
|
|
8
|
-
- Access to various tools like calculators, web - search, etc.
|
|
9
|
-
- Support for multiple platforms
|
|
10
|
-
|
|
11
|
-
### Architecture
|
|
12
|
-
The core components of Cognitrix are:
|
|
13
|
-
|
|
14
|
-
- `Agent` - Base class for chatbot agents with - support for tools
|
|
15
|
-
- `LLM` - Integration with large language models like - Cohere, GPT-3 etc.
|
|
16
|
-
- `Tools` - Various utility tools like calculators, search etc.
|
|
17
|
-
- `Memory` - For tracking context and state
|
|
18
|
-
|
|
19
|
-
The architecture is highly modular and extensible. New tools and capabilities can be easily added.
|
|
20
|
-
|
|
21
|
-
### Getting Started
|
|
22
|
-
|
|
23
|
-
### Installation
|
|
24
|
-
```bash
|
|
25
|
-
pip install cognitrix
|
|
26
|
-
```
|
|
27
|
-
or
|
|
28
|
-
```bash
|
|
29
|
-
pip install https://github.com/theonlyamos/cognitrix/archive/main.zip
|
|
30
|
-
```
|
|
31
|
-
|
|
32
|
-
### Usage
|
|
33
|
-
Run with default settings
|
|
34
|
-
```bash
|
|
35
|
-
cognitrix
|
|
36
|
-
```
|
|
37
|
-
|
|
38
|
-
List supported platforms
|
|
39
|
-
```bash
|
|
40
|
-
cognitrix --platforms
|
|
41
|
-
```
|
|
42
|
-
|
|
43
|
-
Run with specific platform
|
|
44
|
-
```bash
|
|
45
|
-
cognitrix --platform <platform_name>
|
|
46
|
-
```
|
|
47
|
-
|
|
48
|
-
Create a new agent
|
|
49
|
-
```bash
|
|
50
|
-
cognitrix agents --new
|
|
51
|
-
```
|
|
52
|
-
|
|
53
|
-
List created agents
|
|
54
|
-
```bash
|
|
55
|
-
cognitrix agents
|
|
56
|
-
```
|
|
57
|
-
|
|
58
|
-
Run with specific agent
|
|
59
|
-
```bash
|
|
60
|
-
cognitrix --agent <agent_name>
|
|
61
|
-
```
|
|
62
|
-
|
|
63
|
-
Print help message
|
|
64
|
-
```bash
|
|
65
|
-
cognitrix --help
|
|
66
|
-
```
|
|
67
|
-
|
|
68
|
-
### Contributing
|
|
69
|
-
Cognitrix is open source and contributions are welcome. See [CONTRIBUTING.md](CONTRIBUTING.md) for more details.
|
|
70
|
-
|
|
71
|
-
### License
|
|
72
|
-
This project is licensed under the MIT license. See [LICENSE.md](LICENSE.md) for more details.
|
|
73
|
-
|
|
74
|
-
### Credits
|
|
75
|
-
Cognitrix was created by [Amos Amissah](https://github.com/theonlyamos).
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|