npcsh 0.3.27.1__tar.gz → 0.3.27.3__tar.gz

{npcsh-0.3.27.1/npcsh.egg-info → npcsh-0.3.27.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: npcsh
-Version: 0.3.27.1
+Version: 0.3.27.3
 Summary: npcsh is a command line tool for integrating LLMs into everyday workflows and for orchestrating teams of NPCs.
 Home-page: https://github.com/cagostino/npcsh
 Author: Christopher Agostino
@@ -10,44 +10,58 @@ Classifier: License :: OSI Approved :: MIT License
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: redis
-Requires-Dist: flask_sse
-Requires-Dist: anthropic
-Requires-Dist: beautifulsoup4
-Requires-Dist: google-generativeai
-Requires-Dist: google-genai
-Requires-Dist: duckduckgo-search
-Requires-Dist: openai
 Requires-Dist: jinja2
+Requires-Dist: scipy
+Requires-Dist: numpy
 Requires-Dist: requests
 Requires-Dist: markdown
 Requires-Dist: PyYAML
-Requires-Dist: langchain
-Requires-Dist: langchain_community
 Requires-Dist: pygments
 Requires-Dist: termcolor
 Requires-Dist: colorama
+Requires-Dist: Pillow
 Requires-Dist: python-dotenv
-Requires-Dist: pytest
-Requires-Dist: googlesearch-python
+Requires-Dist: pandas
+Requires-Dist: beautifulsoup4
+Requires-Dist: duckduckgo-search
 Requires-Dist: flask
 Requires-Dist: flask_cors
-Requires-Dist: librosa
-Requires-Dist: pandas
-Requires-Dist: matplotlib
-Requires-Dist: IPython
-Requires-Dist: pyautogui
-Requires-Dist: nltk
-Requires-Dist: thefuzz
-Requires-Dist: pypdf
-Requires-Dist: PyMuPDF
-Requires-Dist: screeninfo
-Requires-Dist: sentence_transformers
-Requires-Dist: opencv-python
-Requires-Dist: ollama
-Requires-Dist: kuzu
-Requires-Dist: chromadb
-Requires-Dist: diffusers
+Requires-Dist: redis
+Requires-Dist: flask_sse
+Provides-Extra: lite
+Requires-Dist: anthropic; extra == "lite"
+Requires-Dist: openai; extra == "lite"
+Requires-Dist: google-generativeai; extra == "lite"
+Requires-Dist: google-genai; extra == "lite"
+Provides-Extra: local
+Requires-Dist: sentence_transformers; extra == "local"
+Requires-Dist: opencv-python; extra == "local"
+Requires-Dist: ollama; extra == "local"
+Requires-Dist: kuzu; extra == "local"
+Requires-Dist: chromadb; extra == "local"
+Requires-Dist: diffusers; extra == "local"
+Provides-Extra: whisper
+Requires-Dist: openai-whisper; extra == "whisper"
+Requires-Dist: pyaudio; extra == "whisper"
+Requires-Dist: gtts; extra == "whisper"
+Requires-Dist: playsound==1.2.2; extra == "whisper"
+Requires-Dist: pyttsx3; extra == "whisper"
+Provides-Extra: all
+Requires-Dist: anthropic; extra == "all"
+Requires-Dist: openai; extra == "all"
+Requires-Dist: google-generativeai; extra == "all"
+Requires-Dist: google-genai; extra == "all"
+Requires-Dist: sentence_transformers; extra == "all"
+Requires-Dist: opencv-python; extra == "all"
+Requires-Dist: ollama; extra == "all"
+Requires-Dist: kuzu; extra == "all"
+Requires-Dist: chromadb; extra == "all"
+Requires-Dist: diffusers; extra == "all"
+Requires-Dist: openai-whisper; extra == "all"
+Requires-Dist: pyaudio; extra == "all"
+Requires-Dist: gtts; extra == "all"
+Requires-Dist: playsound==1.2.2; extra == "all"
+Requires-Dist: pyttsx3; extra == "all"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier
@@ -55,6 +69,7 @@ Dynamic: description
 Dynamic: description-content-type
 Dynamic: home-page
 Dynamic: license-file
+Dynamic: provides-extra
 Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary

{npcsh-0.3.27.1 → npcsh-0.3.27.3}/npcsh/search.py

@@ -5,7 +5,11 @@ import os
 
 from bs4 import BeautifulSoup
 from duckduckgo_search import DDGS
-from googlesearch import search
+
+try:
+    from googlesearch import search
+except:
+    pass
 from typing import List, Dict, Any, Optional, Union
 import numpy as np
 import json

{npcsh-0.3.27.1 → npcsh-0.3.27.3/npcsh.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: npcsh
-Version: 0.3.27.1
+Version: 0.3.27.3
 Summary: npcsh is a command line tool for integrating LLMs into everyday workflows and for orchestrating teams of NPCs.
 Home-page: https://github.com/cagostino/npcsh
 Author: Christopher Agostino
@@ -10,44 +10,58 @@ Classifier: License :: OSI Approved :: MIT License
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: redis
-Requires-Dist: flask_sse
-Requires-Dist: anthropic
-Requires-Dist: beautifulsoup4
-Requires-Dist: google-generativeai
-Requires-Dist: google-genai
-Requires-Dist: duckduckgo-search
-Requires-Dist: openai
 Requires-Dist: jinja2
+Requires-Dist: scipy
+Requires-Dist: numpy
 Requires-Dist: requests
 Requires-Dist: markdown
 Requires-Dist: PyYAML
-Requires-Dist: langchain
-Requires-Dist: langchain_community
 Requires-Dist: pygments
 Requires-Dist: termcolor
 Requires-Dist: colorama
+Requires-Dist: Pillow
 Requires-Dist: python-dotenv
-Requires-Dist: pytest
-Requires-Dist: googlesearch-python
+Requires-Dist: pandas
+Requires-Dist: beautifulsoup4
+Requires-Dist: duckduckgo-search
 Requires-Dist: flask
 Requires-Dist: flask_cors
-Requires-Dist: librosa
-Requires-Dist: pandas
-Requires-Dist: matplotlib
-Requires-Dist: IPython
-Requires-Dist: pyautogui
-Requires-Dist: nltk
-Requires-Dist: thefuzz
-Requires-Dist: pypdf
-Requires-Dist: PyMuPDF
-Requires-Dist: screeninfo
-Requires-Dist: sentence_transformers
-Requires-Dist: opencv-python
-Requires-Dist: ollama
-Requires-Dist: kuzu
-Requires-Dist: chromadb
-Requires-Dist: diffusers
+Requires-Dist: redis
+Requires-Dist: flask_sse
+Provides-Extra: lite
+Requires-Dist: anthropic; extra == "lite"
+Requires-Dist: openai; extra == "lite"
+Requires-Dist: google-generativeai; extra == "lite"
+Requires-Dist: google-genai; extra == "lite"
+Provides-Extra: local
+Requires-Dist: sentence_transformers; extra == "local"
+Requires-Dist: opencv-python; extra == "local"
+Requires-Dist: ollama; extra == "local"
+Requires-Dist: kuzu; extra == "local"
+Requires-Dist: chromadb; extra == "local"
+Requires-Dist: diffusers; extra == "local"
+Provides-Extra: whisper
+Requires-Dist: openai-whisper; extra == "whisper"
+Requires-Dist: pyaudio; extra == "whisper"
+Requires-Dist: gtts; extra == "whisper"
+Requires-Dist: playsound==1.2.2; extra == "whisper"
+Requires-Dist: pyttsx3; extra == "whisper"
+Provides-Extra: all
+Requires-Dist: anthropic; extra == "all"
+Requires-Dist: openai; extra == "all"
+Requires-Dist: google-generativeai; extra == "all"
+Requires-Dist: google-genai; extra == "all"
+Requires-Dist: sentence_transformers; extra == "all"
+Requires-Dist: opencv-python; extra == "all"
+Requires-Dist: ollama; extra == "all"
+Requires-Dist: kuzu; extra == "all"
+Requires-Dist: chromadb; extra == "all"
+Requires-Dist: diffusers; extra == "all"
+Requires-Dist: openai-whisper; extra == "all"
+Requires-Dist: pyaudio; extra == "all"
+Requires-Dist: gtts; extra == "all"
+Requires-Dist: playsound==1.2.2; extra == "all"
+Requires-Dist: pyttsx3; extra == "all"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier
@@ -55,6 +69,7 @@ Dynamic: description
 Dynamic: description-content-type
 Dynamic: home-page
 Dynamic: license-file
+Dynamic: provides-extra
 Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary

{npcsh-0.3.27.1 → npcsh-0.3.27.3}/npcsh.egg-info/SOURCES.txt

@@ -64,4 +64,5 @@ tests/test_npc_compiler.py
 tests/test_npcsh.py
 tests/test_npcteam.py
 tests/test_shell_helpers.py
+tests/test_tars.py
 tests/test_tool_use.py

npcsh-0.3.27.3/npcsh.egg-info/requires.txt

@@ -0,0 +1,56 @@
+jinja2
+scipy
+numpy
+requests
+markdown
+PyYAML
+pygments
+termcolor
+colorama
+Pillow
+python-dotenv
+pandas
+beautifulsoup4
+duckduckgo-search
+flask
+flask_cors
+redis
+flask_sse
+
+[all]
+anthropic
+openai
+google-generativeai
+google-genai
+sentence_transformers
+opencv-python
+ollama
+kuzu
+chromadb
+diffusers
+openai-whisper
+pyaudio
+gtts
+playsound==1.2.2
+pyttsx3
+
+[lite]
+anthropic
+openai
+google-generativeai
+google-genai
+
+[local]
+sentence_transformers
+opencv-python
+ollama
+kuzu
+chromadb
+diffusers
+
+[whisper]
+openai-whisper
+pyaudio
+gtts
+playsound==1.2.2
+pyttsx3

{npcsh-0.3.27.1 → npcsh-0.3.27.3}/setup.py

@@ -1,17 +1,8 @@
-# Force lite installation if environment variable is set
-import os
-
-import sys
-
-if os.environ.get("NPCSH_LITE_INSTALL", "").lower() == "true":
-
-    sys.argv.append("--config-settings")
-    sys.argv.append("install.lite=true")
-
 from setuptools import setup, find_packages
 import site
 import platform
 from pathlib import Path
+import os
 
 
 def package_files(directory):
@@ -58,47 +49,36 @@ npcsh-setup
 To configure your API keys and preferences.
 ==============================================
 """
-    return ""  # Return empty string for non-Windows platforms
+    return ""
 
 
-# Define core (lite) requirements
-core_requirements = [
-    "redis",
-    "flask_sse",
-    "anthropic",
-    "beautifulsoup4",
-    "google-generativeai",
-    "google-genai",
-    "duckduckgo-search",
-    "openai",
+# Base requirements (no LLM packages)
+base_requirements = [
     "jinja2",
+    "scipy",
+    "numpy",
     "requests",
     "markdown",
     "PyYAML",
-    "langchain",
-    "langchain_community",
     "pygments",
     "termcolor",
     "colorama",
+    "Pillow",
     "python-dotenv",
-    "pytest",
-    "googlesearch-python",
+    "pandas",
+    "beautifulsoup4",
+    "duckduckgo-search",
     "flask",
     "flask_cors",
-    "librosa",
-    "pandas",
-    "matplotlib",
-    "IPython",
-    "pyautogui",
-    "nltk",
-    "thefuzz",
-    "pypdf",
-    "PyMuPDF",
-    "screeninfo",
+    "redis",
+    "flask_sse",
 ]
 
-# Define additional requirements for full installation
-extra_requirements = [
+# API integration requirements
+api_requirements = ["anthropic", "openai", "google-generativeai", "google-genai"]
+
+# Local ML/AI requirements
+local_requirements = [
     "sentence_transformers",
     "opencv-python",
     "ollama",
@@ -107,8 +87,8 @@ extra_requirements = [
     "diffusers",
 ]
 
-# Define audio requirements
-audio_requirements = [
+# Voice/Audio requirements
+voice_requirements = [
     "openai-whisper",
     "pyaudio",
     "gtts",
@@ -118,21 +98,17 @@ audio_requirements = [
 
 extra_files = package_files("npcsh/npc_team/")
 
-
-def get_requirements():
-    # Check if lite installation was requested via sys.argv
-    if any("install.lite=true" in arg for arg in sys.argv):
-        return core_requirements
-    elif os.environ.get("NPCSH_AUDIO_INSTALL", "").lower() == "true":
-        return core_requirements + extra_requirements + audio_requirements
-    return core_requirements + extra_requirements
-
-
 setup(
     name="npcsh",
-    version="0.3.27.1",
+    version="0.3.27.3",
     packages=find_packages(exclude=["tests*"]),
-    install_requires=get_requirements(),
+    install_requires=base_requirements,  # Only install base requirements by default
+    extras_require={
+        "lite": api_requirements,  # Just API integrations
+        "local": local_requirements,  # Local AI/ML features
+        "whisper": voice_requirements,  # Voice/Audio features
+        "all": api_requirements + local_requirements + voice_requirements,  # Everything
+    },
     entry_points={
         "console_scripts": [
             "npcsh=npcsh.shell:main",
@@ -154,6 +130,5 @@ setup(
     python_requires=">=3.10",
 )
 
-# Print setup message only on Windows
 if platform.system() == "Windows":
     print(get_setup_message())

npcsh-0.3.27.3/tests/test_tars.py

@@ -0,0 +1,178 @@
+import os
+import base64
+import requests
+import json
+import torch
+import pyautogui  # For taking screenshots
+import time
+from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessor
+from screeninfo import get_monitors
+from npcsh.llm_funcs import get_openai_response
+
+
+def get_screen_resolution():
+    monitor = get_monitors()[0]  # Get primary monitor
+    return monitor.width, monitor.height
+
+
+from PIL import Image
+
+
+def capture_screenshot() -> str:
+    """Captures a screenshot and saves it to a specified path."""
+    screenshot_path = "screenshot.png"
+    screenshot = pyautogui.screenshot()
+
+    # Resize screenshot to fit model's pixel range
+    desired_width = 1280  # Adjust as needed based on max_pixels range
+    desired_height = int(
+        (desired_width * screenshot.height) / screenshot.width
+    )  # Maintain aspect ratio
+    screenshot = screenshot.resize((desired_width, desired_height))
+
+    screenshot.save(screenshot_path)
+    return screenshot_path
+
+
+# Adjust processor for specific pixel range
+min_pixels = 256 * 28 * 28
+max_pixels = 1280 * 28 * 28
+
+
+def encode_image_to_base64(image_path: str) -> str:
+    """Encodes an image file to a base64 string."""
+    with open(image_path, "rb") as image_file:
+        encoded_string = base64.b64encode(image_file.read()).decode("utf-8")
+    return f"data:image/png;base64,{encoded_string}"
+
+
+def get_tars_response(command: str, model_name: str) -> str:
+    """Generates a response from the UI-TARS model based on the command and screenshot image."""
+    # model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
+    #    model_name, torch_dtype="auto", device_map="auto"
+    # )
+    # processor = AutoProcessor.from_pretrained(model_name)
+
+    # capture the current screen
+    im = capture_screenshot()
+    image_data = encode_image_to_base64(im)
+    prompt = (
+        f"""You are a GUI agent. You are given a task and your action history,
+    with screenshots.     You need to perform the next action or set of actions to complete the task.
+    here is the task you must complete: {command}
+        """
+        + r"""
+    click(start_box='<|box_start|>(x1,y1)<|box_end|>')
+    left_double(start_box='<|box_start|>(x1,y1)<|box_end|>')
+    right_single(start_box='<|box_start|>(x1,y1)<|box_end|>')
+    drag(start_box='<|box_start|>(x1,y1)<|box_end|>', end_box='<|box_start|>(x3,y3)<|box_end|>')
+    hotkey(key='')
+    type(content='') #If you want to submit your input, use "\
+    " at the end of `content`.
+    scroll(start_box='<|box_start|>(x1,y1)<|box_end|>', direction='down or up or right or left')
+    wait() #Sleep for 5s and take a screenshot to check for any changes.
+    finished()
+    call_user() # Submit the task and call the user when the task is unsolvable, or when you need the user's help.
+
+    your response should be a list of actions to perform in the order they should be performed.
+    Provide a single json object with the following format:
+    { "actions":  ['action1', 'action2', 'action3']    }
+    Do not provide any additional text or markdown formatting.
+    """
+    )
+
+    messages = [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "image_url",
+                    "image_url": {"url": image_data},
+                },
+                {"type": "text", "text": command},
+            ],
+        }
+    ]
+
+    # tars:
+    """text = processor.apply_chat_template(
+        messages, tokenize=False, add_generation_prompt=True
+    )
+    inputs = processor(text=text, padding=True, return_tensors="pt").to(model.device)
+    generated_ids = model.generate(**inputs, max_new_tokens=128)
+    output_text = processor.batch_decode(generated_ids, skip_special_tokens=True)
+    return output_text[0]
+    """
+    gpt4o_response = get_openai_response(
+        prompt, model="gpt-4o-mini", messages=messages, format="json"
+    )
+
+    return gpt4o_response
+
+
+def execute_actions(actions: list):
+    """Executes the actions received from the model using pyautogui."""
+    for action in actions:
+        if action.startswith("click"):
+            x, y = map(int, action[action.find("(") + 1 : action.find(")")].split(","))
+            pyautogui.click(x, y)
+        elif action.startswith("left_double"):
+            x, y = map(int, action[action.find("(") + 1 : action.find(")")].split(","))
+            pyautogui.doubleClick(x, y)
+        elif action.startswith("right_single"):
+            x, y = map(int, action[action.find("(") + 1 : action.find(")")].split(","))
+            pyautogui.rightClick(x, y)
+        elif action.startswith("drag"):
+            coords = list(
+                map(
+                    int,
+                    action[action.find("(") + 1 : action.find(")")]
+                    .replace("(", "")
+                    .replace(")", "")
+                    .split(","),
+                )
+            )
+            pyautogui.moveTo(coords[0], coords[1])
+            pyautogui.dragTo(coords[2], coords[3], duration=0.5)
+        elif action.startswith("type"):
+            text = action.split("('")[1].split("')")[0]
+            pyautogui.write(text, interval=0.05)
+        elif action.startswith("hotkey"):
+            key = action.split("('")[1].split("')")[0]
+            pyautogui.hotkey(key)
+        elif action.startswith("scroll"):
+            direction = action.split("('")[1].split("')")[0]
+            amount = -100 if direction == "down" else 100
+            pyautogui.scroll(amount)
+        elif action.startswith("wait"):
+            time.sleep(5)
+        elif action.startswith("finished"):
+            print("Task completed.")
+
+
+def ui_tars_control_loop(model_name: str):
+    """Main loop for interacting with the user and executing commands via UI-TARS."""
+    print("UI-TARS Control Loop Started.")
+    screen_width, screen_height = get_screen_resolution()
+    print(f"Screen resolution: {screen_width}x{screen_height}")
+
+    while True:
+        command = input("Enter your command (or type 'exit' to quit): ")
+        if command.lower() == "exit":
+            print("Exiting UI-TARS Control Loop.")
+            break
+
+        screenshot_path = capture_screenshot()
+        tars_result = get_tars_response(command, screenshot_path, model_name)
+        print(f"UI-TARS Response: {tars_result}")
+
+        try:
+            actions = json.loads(tars_result).get("actions", [])
+            execute_actions(actions)
+        except json.JSONDecodeError:
+            print("Error parsing actions from UI-TARS response.")
+
+
+if __name__ == "__main__":
+    MODEL_NAME = "ui-tars-7B"  # Replace with your actual UI-TARS model name
+    ui_tars_control_loop(MODEL_NAME)

npcsh-0.3.27.1/npcsh.egg-info/requires.txt

@@ -1,38 +0,0 @@
-redis
-flask_sse
-anthropic
-beautifulsoup4
-google-generativeai
-google-genai
-duckduckgo-search
-openai
-jinja2
-requests
-markdown
-PyYAML
-langchain
-langchain_community
-pygments
-termcolor
-colorama
-python-dotenv
-pytest
-googlesearch-python
-flask
-flask_cors
-librosa
-pandas
-matplotlib
-IPython
-pyautogui
-nltk
-thefuzz
-pypdf
-PyMuPDF
-screeninfo
-sentence_transformers
-opencv-python
-ollama
-kuzu
-chromadb
-diffusers