You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
### **Compiled versions for Linux and Windows**: https://pygpt.net/#download
8
8
9
9
## Overview
10
10
11
-
**PYGPT** is **all-in-one**desktop AI assistant that provides direct interaction with OpenAI language models, including `GPT-4`, `GPT-4 Vision`, and `GPT-3.5`, through the `OpenAI API`. The application also integrates with alternative LLMs, like those available on `HuggingFace`, by utilizing `Langchain`.
11
+
**PYGPT** is **all-in-one**Desktop AI Assistant that provides direct interaction with OpenAI language models, including `GPT-4`, `GPT-4 Vision`, and `GPT-3.5`, through the `OpenAI API`. The application also integrates with alternative LLMs, like those available on `HuggingFace`, by utilizing `Langchain`.
12
12
13
-
This assistant offers multiple modes of operation such as chat, assistants, completions, and image-related tasks using `DALL-E 3` for generation and `GPT-4 Vision` for analysis. **PYGPT** has filesystem capabilities for file I/O, can generate and run Python code, execute system commands, and manage file transfers. It also allows models to perform web searches with the `Google Custom Search API`.
13
+
This assistant offers multiple modes of operation such as chat, assistants, completions, and image-related tasks using `DALL-E 3` for generation and `GPT-4 Vision` for analysis. **PYGPT** has filesystem capabilities for file I/O, can generate and run Python code, execute system commands, execute custom commands and manage file transfers. It also allows models to perform web searches with the `Google Custom Search API`.
14
14
15
15
For audio interactions, **PYGPT** includes speech synthesis using the `Microsoft Azure Text-to-Speech API` and `OpenAI's TTS API`. Additionally, it features speech recognition capabilities provided by `OpenAI Whisper`, enabling the application to understand spoken commands and transcribe audio inputs into text. It features context memory with save and load functionality, enabling users to resume interactions from predefined points in the conversation. Prompt creation and management are streamlined through an intuitive preset system.
16
16
@@ -35,6 +35,7 @@ You can download compiled version for Windows and Linux here: https://pygpt.net/
35
35
- Image analysis via `GPT-4 Vision`.
36
36
- Integrated `Langchain` support (you can connect to any LLM, e.g., on `HuggingFace`).
37
37
- Commands execution (via plugins: access to the local filesystem, Python code interpreter, system commands execution).
38
+
- Custom commands creation and execution
38
39
- Manages files and attachments with options to upload, download, and organize.
39
40
- Context history with the capability to revert to previous contexts (long-term memory).
40
41
- Allows you to easily manage prompts with handy editable presets.
@@ -370,6 +371,8 @@ The plugin can also execute system commands, allowing GPT to integrate with your
370
371
Plugins can work in conjunction to perform sequential tasks; for example, the `Files` plugin can write generated
371
372
Python code to a file, which the `Code Interpreter` can execute it and return its result to GPT.
372
373
374
+
-`Command: Custom Commands` - allows you to create and execute custom commands on your system.
375
+
373
376
-`Audio Output (Microsoft Azure)` - provides voice synthesis using the Microsoft Azure Text To Speech API.
374
377
375
378
-`Audio Output (OpenAI TTS)` - provides voice synthesis using the OpenAI Text To Speech API.
@@ -522,10 +525,11 @@ The plugin allows for file management within the local filesystem. It enables th
522
525
Plugin capabilities include:
523
526
524
527
- Reading files
528
+
- Appending to files
525
529
- Writing files
526
-
- Executing code from files
527
-
- Creating directories
530
+
- Deleting files
528
531
- Listing files and directories
532
+
- Creating directories
529
533
530
534
If a file being created (with the same name) already exists, a prefix including the date and time is added to the file name.
With the `Custom Commands` plugin, you can integrate PYGPT with your operating system and scripts or applications. You can define an unlimited number of custom commands and instruct GPT on when and how to execute them. Configuration is straightforward, and PYGPT includes a simple tutorial command for testing and learning how it works:
To add a new custom command, click the 'ADD' button and then:
603
+
604
+
1. Provide a name for your command: this is a unique identifier for GPT.
605
+
2. Provide an `instruction` explaining what this command does; GPT will know when to use the command based on this instruction.
606
+
3. Define `params`, separated by commas - GPT will send data to your commands using these params. These params will be placed into placeholders you have defined in the `cmd` field. For example:
607
+
608
+
If you want instruct GPT to execute your Python script named `smart_home_lights.py` with an argument, such as `1` to turn the light ON, and `0` to turn it OFF, define it as follows:
609
+
610
+
-**name**: lights_cmd
611
+
-**instruction**: turn lights on/off; use 1 as 'arg' to turn ON, or 0 as 'arg' to turn OFF
When you ask GPT to turn your lights ON, GPT will locate this command and prepare the command `python /path/to/smart_home_lights.py {arg}` with `{arg}` replaced with `1`. On your system, it will execute the command:
618
+
619
+
```python /path/to/smart_home_lights.py 1```
620
+
621
+
And that's all. GPT will take care of the rest when you ask to turn ON the lights.
622
+
623
+
You can define as many placeholders and parameters as you desire.
624
+
625
+
Here are some predefined system placeholders for use:
626
+
627
+
-`{_time}` - current time in `Y-m-d` format
628
+
-`{_date}` - current date in `H:M:S` format
629
+
-`{_datetime}` - current date and time in `Y-m-d H:M:S` format
630
+
-`{_file}` - path to the file from which the command is invoked
631
+
-`{_home}` - path to PYGPT's home/workworking directory
632
+
633
+
You can connect predefined placeholders with your own params.
634
+
635
+
*Example:*
636
+
637
+
-**name**: song_cmd
638
+
-**instruction**: store the generated song on hard disk
With the setup above, every time you ask GPT to generate a song for you and save it to the disk, it will:
644
+
645
+
1. Generate a song.
646
+
2. Locate your command.
647
+
3. Execute the command by sending the song's title and text.
648
+
4. The command will save the song text into a file named with the song's title in the PYGPT working directory.
649
+
650
+
**Example tutorial command**
651
+
652
+
PYGPT provides simple tutorial command to show how it work, to run it just ask GPT for execute `tutorial test command` and it will show you how it works:
Copy file name to clipboardExpand all lines: docs/source/intro.rst
+2-1Lines changed: 2 additions & 1 deletion
Original file line number
Diff line number
Diff line change
@@ -4,7 +4,7 @@ Introduction
4
4
Overview
5
5
----------------
6
6
7
-
**PYGPT** is **all-in-one** desktop AI assistant that provides direct interaction with OpenAI language models, including ``GPT-4``, ``GPT-4 Vision``, and ``GPT-3.5``, through the ``OpenAI API``. The application also integrates with alternative LLMs, like those available on ``HuggingFace``, by utilizing ``Langchain``.
7
+
**PYGPT** is **all-in-one** Desktop AI Assistant that provides direct interaction with OpenAI language models, including ``GPT-4``, ``GPT-4 Vision``, and ``GPT-3.5``, through the ``OpenAI API``. The application also integrates with alternative LLMs, like those available on ``HuggingFace``, by utilizing ``Langchain``.
8
8
9
9
This assistant offers multiple modes of operation such as chat, assistants, completions, and image-related tasks using ``DALL-E 3`` for generation and ``GPT-4 Vision`` for analysis. **PYGPT** has filesystem capabilities for file I/O, can generate and run Python code, execute system commands, and manage file transfers. It also allows models to perform web searches with the ``Google Custom Search API``.
10
10
@@ -30,6 +30,7 @@ Features
30
30
* Image analysis via ``GPT-4 Vision``.
31
31
* Integrated ``Langchain`` support (you can connect to any LLM, e.g., on ``HuggingFace``).
32
32
* Commands execution (via plugins: access to the local filesystem, Python code interpreter, system commands execution).
33
+
* Custom commands creation and execution
33
34
* Manages files and attachments with options to upload, download, and organize.
34
35
* Context history with the capability to revert to previous contexts (long-term memory).
35
36
* Allows you to easily manage prompts with handy editable presets.
Copy file name to clipboardExpand all lines: docs/source/plugins.rst
+1Lines changed: 1 addition & 0 deletions
Original file line number
Diff line number
Diff line change
@@ -8,6 +8,7 @@ The following plugins are currently available, and GPT can use them instantly:
8
8
* ``Command: Google Web Search`` - allows searching the internet via the `Google Custom Search Engine`.
9
9
* ``Command: Files I/O`` - grants access to the local filesystem, enabling GPT to read and write files, as well as list and create directories.
10
10
* ``Command: Code Interpreter`` - responsible for generating and executing Python code, functioning much like the `Code Interpreter` on `ChatGPT`, but locally. This means GPT can interface with any script, application, or code. The plugin can also execute system commands, allowing GPT to integrate with your operating system. Plugins can work in conjunction to perform sequential tasks; for example, the `Files` plugin can write generated Python code to a file, which the `Code Interpreter` can execute it and return its result to GPT.
11
+
* ``Command: Custom Commands`` - allows you to create and execute custom commands on your system.
11
12
* ``Audio Output (Microsoft Azure)`` - provides voice synthesis using the Microsoft Azure Text To Speech API.
12
13
* ``Audio Output (OpenAI TTS)`` - provides voice synthesis using the `OpenAI Text To Speech API`.
13
14
* ``Audio Input (OpenAI Whisper)`` - offers speech recognition through the `OpenAI Whisper API`.
Allow system commands execution. *Default:* `True`
78
79
79
80
81
+
Command: Custom Commands
82
+
------------------------
83
+
84
+
With the ``Custom Commands`` plugin, you can integrate **PYGPT** with your operating system and scripts or applications. You can define an unlimited number of custom commands and instruct GPT on when and how to execute them. Configuration is straightforward, and **PYGPT** includes a simple tutorial command for testing and learning how it works:
85
+
86
+
.. image:: images/v2_custom_cmd.png
87
+
:width:800
88
+
89
+
To add a new custom command, click the **ADD** button and then:
90
+
91
+
1. Provide a name for your command: this is a unique identifier for GPT.
92
+
2. Provide an ``instruction`` explaining what this command does; GPT will know when to use the command based on this instruction.
93
+
3. Define ``params``, separated by commas - GPT will send data to your commands using these params. These params will be placed into placeholders you have defined in the ``cmd`` field. For example:
94
+
95
+
If you want instruct GPT to execute your Python script named ``smart_home_lights.py`` with an argument, such as ``1`` to turn the light ON, and ``0`` to turn it OFF, define it as follows:
96
+
97
+
- **name**: lights_cmd
98
+
- **instruction**: turn lights on/off; use 1 as 'arg' to turn ON, or 0 as 'arg' to turn OFF
When you ask GPT to turn your lights ON, GPT will locate this command and prepare the command ``python /path/to/smart_home_lights.py {arg}`` with ``{arg}`` replaced with ``1``. On your system, it will execute the command:
105
+
106
+
.. code-block:: console
107
+
108
+
python /path/to/smart_home_lights.py 1
109
+
110
+
And that's all. GPT will take care of the rest when you ask to turn ON the lights.
111
+
112
+
You can define as many placeholders and parameters as you desire.
113
+
114
+
Here are some predefined system placeholders for use:
115
+
116
+
- ``{_time}`` - current time in ``Y-m-d`` format
117
+
- ``{_date}`` - current date in ``H:M:S`` format
118
+
- ``{_datetime}`` - current date and time in ``Y-m-d H:M:S`` format
119
+
- ``{_file}`` - path to the file from which the command is invoked
120
+
- ``{_home}`` - path to PYGPT's home/working directory
121
+
122
+
You can connect predefined placeholders with your own params.
123
+
124
+
*Example:*
125
+
126
+
- **name**: song_cmd
127
+
- **instruction**: store the generated song on hard disk
With the setup above, every time you ask GPT to generate a song for you and save it to the disk, it will:
133
+
134
+
1. Generate a song.
135
+
2. Locate your command.
136
+
3. Execute the command by sending the song's title and text.
137
+
4. The command will save the song text into a file named with the song's title in the PYGPT working directory.
138
+
139
+
**Example tutorial command**
140
+
141
+
**PYGPT** provides simple tutorial command to show how it work, to run it just ask GPT for execute ``tutorial test command`` and it will show you how it works:
0 commit comments