orgo 0.0.13__tar.gz → 0.0.15__tar.gz

{orgo-0.0.13 → orgo-0.0.15}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: orgo
-Version: 0.0.13
+Version: 0.0.15
 Summary: Computers for AI agents
 Author: Orgo Team
 License: MIT
@@ -35,6 +35,9 @@ computer.type("Hello world")
 computer.key("Enter")
 computer.screenshot()  # Returns PIL Image
 
+# Execute Python code
+computer.exec("import pyautogui; pyautogui.click(512, 384)")
+
 # Cleanup
 computer.shutdown()
 ```

{orgo-0.0.13 → orgo-0.0.15}/README.md

@@ -22,6 +22,9 @@ computer.type("Hello world")
 computer.key("Enter")
 computer.screenshot()  # Returns PIL Image
 
+# Execute Python code
+computer.exec("import pyautogui; pyautogui.click(512, 384)")
+
 # Cleanup
 computer.shutdown()
 ```

{orgo-0.0.13 → orgo-0.0.15}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "orgo"
-version = "0.0.13"
+version = "0.0.15"
 description = "Computers for AI agents"
 authors = [{name = "Orgo Team"}]
 license = {text = "MIT"}

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/__init__.py

@@ -1,3 +1,4 @@
+# src/orgo/__init__.py
 """Orgo SDK: Desktop infrastructure for AI agents"""
 
 from .computer import Computer

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/api/__init__.py

@@ -1,3 +1,4 @@
+# src/orgo/api/__init__.py
 """API package for Orgo SDK"""
 
 from .client import ApiClient

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/api/client.py

@@ -1,5 +1,5 @@
-"""API client for Orgo service"""
 # src/orgo/api/client.py
+"""API client for Orgo service"""
 
 import requests
 from typing import Dict, Any, Optional
@@ -96,6 +96,13 @@ class ApiClient:
             "command": command
         })
     
+    def execute_python(self, project_id: str, code: str, timeout: int = 10) -> Dict[str, Any]:
+        """Execute Python code on the computer"""
+        return self._request("POST", f"computers/{project_id}/exec", {
+            "code": code,
+            "timeout": timeout
+        })
+    
     def wait(self, project_id: str, seconds: float) -> Dict[str, Any]:
         return self._request("POST", f"computers/{project_id}/wait", {
             "seconds": seconds

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/computer.py

@@ -1,10 +1,12 @@
+# src/orgo/computer.py
 """Computer class for interacting with Orgo virtual environments"""
 import os
 import io
 import base64
 import logging
-from typing import Dict, List, Any, Optional, Callable, Union
+from typing import Dict, List, Any, Optional, Callable
 from PIL import Image
+import requests
 from requests.exceptions import RequestException
 
 from .api.client import ApiClient
@@ -96,13 +98,33 @@ class Computer:
     def screenshot(self) -> Image.Image:
         """Capture screenshot and return as PIL Image"""
         response = self.api.get_screenshot(self.project_id)
-        img_data = base64.b64decode(response.get("image", ""))
-        return Image.open(io.BytesIO(img_data))
+        image_data = response.get("image", "")
+        
+        # Check if it's a URL (new format) or base64 (legacy format)
+        if image_data.startswith(('http://', 'https://')):
+            # Download image from URL
+            img_response = requests.get(image_data)
+            img_response.raise_for_status()
+            return Image.open(io.BytesIO(img_response.content))
+        else:
+            # Legacy base64 format
+            img_data = base64.b64decode(image_data)
+            return Image.open(io.BytesIO(img_data))
     
     def screenshot_base64(self) -> str:
         """Capture screenshot and return as base64 string"""
         response = self.api.get_screenshot(self.project_id)
-        return response.get("image", "")
+        image_data = response.get("image", "")
+        
+        # Check if it's a URL (new format) or base64 (legacy format)
+        if image_data.startswith(('http://', 'https://')):
+            # Download image from URL and convert to base64
+            img_response = requests.get(image_data)
+            img_response.raise_for_status()
+            return base64.b64encode(img_response.content).decode('utf-8')
+        else:
+            # Already base64
+            return image_data
     
     # Execution methods
     def bash(self, command: str) -> str:
@@ -110,6 +132,37 @@ class Computer:
         response = self.api.execute_bash(self.project_id, command)
         return response.get("output", "")
     
+    def exec(self, code: str, timeout: int = 10) -> Dict[str, Any]:
+        """
+        Execute Python code on the remote computer.
+        
+        Args:
+            code: Python code to execute
+            timeout: Maximum execution time in seconds (default: 10, max: 300)
+            
+        Returns:
+            Dict with keys:
+            - success: bool indicating if execution completed without errors
+            - output: str containing stdout output
+            - error: str containing error message if any
+            - error_type: str with exception type name if error occurred
+            - timeout: bool indicating if execution timed out
+            
+        Example:
+            result = computer.exec('''
+import os
+print(f"Current directory: {os.getcwd()}")
+print(f"Files: {os.listdir('.')}")
+            ''')
+            
+            if result['success']:
+                print(result['output'])
+            else:
+                print(f"Error: {result['error']}")
+        """
+        response = self.api.execute_python(self.project_id, code, timeout)
+        return response
+    
     def wait(self, seconds: float) -> Dict[str, Any]:
         """Wait for specified number of seconds"""
         return self.api.wait(self.project_id, seconds)
@@ -168,4 +221,4 @@ class Computer:
             # Pass through the Orgo API client configuration
             orgo_api_key=self.api_key,
             orgo_base_url=self.base_api_url
-        )
+        )

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/project.py

@@ -1,3 +1,4 @@
+# src/orgo/project.py
 """Project management for Orgo virtual environments"""
 import os
 import json

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/prompt.py

@@ -1,3 +1,4 @@
+# src/orgo/prompt.py
 """
 Prompt module for interacting with virtual computers using AI models.
 """

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/utils/__init__.py

@@ -1,3 +1,4 @@
+# src/orgo/utils/__init__.py
 """Utility functions for Orgo SDK"""
 
 from .auth import get_api_key

{orgo-0.0.13 → orgo-0.0.15}/src/orgo/utils/auth.py

@@ -1,3 +1,4 @@
+# src/orgo/utils/auth.py
 """Authentication utilities for Orgo SDK"""
 
 import os

{orgo-0.0.13 → orgo-0.0.15}/src/orgo.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: orgo
-Version: 0.0.13
+Version: 0.0.15
 Summary: Computers for AI agents
 Author: Orgo Team
 License: MIT
@@ -35,6 +35,9 @@ computer.type("Hello world")
 computer.key("Enter")
 computer.screenshot()  # Returns PIL Image
 
+# Execute Python code
+computer.exec("import pyautogui; pyautogui.click(512, 384)")
+
 # Cleanup
 computer.shutdown()
 ```

{orgo-0.0.13 → orgo-0.0.15}/src/orgo.egg-info/SOURCES.txt

@@ -9,10 +9,6 @@ src/orgo.egg-info/SOURCES.txt
 src/orgo.egg-info/dependency_links.txt
 src/orgo.egg-info/requires.txt
 src/orgo.egg-info/top_level.txt
-src/orgo/adapters/__init__.py
-src/orgo/adapters/anthropic.py
-src/orgo/adapters/base.py
-src/orgo/adapters/openai.py
 src/orgo/api/__init__.py
 src/orgo/api/client.py
 src/orgo/utils/__init__.py

orgo-0.0.13/src/orgo/adapters/__init__.py

@@ -1,7 +0,0 @@
-"""Adapters for integrating Orgo with AI models"""
-
-from .base import BaseAdapter
-from .anthropic import AnthropicAdapter
-from .openai import OpenAIAdapter
-
-__all__ = ["BaseAdapter", "AnthropicAdapter", "OpenAIAdapter"]

orgo-0.0.13/src/orgo/adapters/anthropic.py

@@ -1,64 +0,0 @@
-"""Adapter for Anthropic's Claude"""
-
-from typing import Dict, Any, Optional
-from .base import BaseAdapter
-
-class AnthropicAdapter(BaseAdapter):
-    def get_tool_definition(self) -> Dict[str, Any]:
-        return {
-            "name": "computer",
-            "description": "Controls a virtual computer to automate tasks",
-            "type": "function",
-            "parameters": {
-                "type": "object",
-                "properties": {
-                    "action": {
-                        "type": "string",
-                        "enum": ["left_click", "right_click", "double_click", "type", "key", "scroll", "screenshot"],
-                        "description": "The action to perform on the computer"
-                    },
-                    "coordinate": {
-                        "type": "array",
-                        "items": {"type": "number"},
-                        "description": "The x,y coordinates for click actions",
-                        "minItems": 2,
-                        "maxItems": 2
-                    },
-                    "text": {
-                        "type": "string",
-                        "description": "The text to type or key to press"
-                    },
-                    "direction": {
-                        "type": "string",
-                        "enum": ["up", "down", "left", "right"],
-                        "description": "The direction to scroll"
-                    },
-                    "amount": {
-                        "type": "number",
-                        "description": "The amount to scroll"
-                    }
-                },
-                "required": ["action"],
-                "additionalProperties": False
-            }
-        }
-    
-    def format_result(self, tool_id: str, output: Optional[str] = None, error: Optional[str] = None) -> Dict[str, Any]:
-        screenshot = self.computer.get_base64()
-        result = {
-            "type": "tool_result",
-            "id": tool_id,
-            "content": {
-                "type": "image",
-                "source": {
-                    "type": "base64",
-                    "media_type": "image/png",
-                    "data": screenshot
-                }
-            }
-        }
-        
-        if error:
-            result["error"] = error
-            
-        return result

orgo-0.0.13/src/orgo/adapters/base.py

@@ -1,14 +0,0 @@
-"""Base adapter for AI model integration"""
-
-from typing import Dict, Any, Optional
-from ..computer import Computer
-
-class BaseAdapter:
-    def __init__(self, computer: Computer):
-        self.computer = computer
-    
-    def get_tool_definition(self) -> Dict[str, Any]:
-        raise NotImplementedError("Subclasses must implement get_tool_definition()")
-    
-    def format_result(self, tool_id: str, output: Optional[str] = None, error: Optional[str] = None) -> Dict[str, Any]:
-        raise NotImplementedError("Subclasses must implement format_result()")

orgo-0.0.13/src/orgo/adapters/openai.py

@@ -1,69 +0,0 @@
-"""Adapter for OpenAI models"""
-
-from typing import Dict, Any, Optional
-from .base import BaseAdapter
-
-class OpenAIAdapter(BaseAdapter):
-    def get_tool_definition(self) -> Dict[str, Any]:
-        return {
-            "type": "function",
-            "function": {
-                "name": "computer",
-                "description": "Controls a virtual computer to automate tasks",
-                "parameters": {
-                    "type": "object",
-                    "properties": {
-                        "action": {
-                            "type": "string",
-                            "enum": ["click", "right_click", "double_click", "type", "key", "scroll", "screenshot"],
-                            "description": "The action to perform on the computer"
-                        },
-                        "x": {
-                            "type": "number",
-                            "description": "The x coordinate for click actions"
-                        },
-                        "y": {
-                            "type": "number",
-                            "description": "The y coordinate for click actions"
-                        },
-                        "text": {
-                            "type": "string",
-                            "description": "The text to type or key to press"
-                        },
-                        "direction": {
-                            "type": "string",
-                            "enum": ["up", "down", "left", "right"],
-                            "description": "The direction to scroll"
-                        },
-                        "amount": {
-                            "type": "number",
-                            "description": "The amount to scroll"
-                        }
-                    },
-                    "required": ["action"],
-                    "additionalProperties": False
-                }
-            }
-        }
-    
-    def format_result(self, tool_id: str, output: Optional[str] = None, error: Optional[str] = None) -> Dict[str, Any]:
-        screenshot = self.computer.get_base64()
-        result = {
-            "tool_call_id": tool_id,
-            "content": [
-                {
-                    "type": "image_url",
-                    "image_url": {
-                        "url": f"data:image/png;base64,{screenshot}"
-                    }
-                }
-            ]
-        }
-        
-        if error:
-            result["content"].insert(0, {
-                "type": "text",
-                "text": f"Error: {error}"
-            })
-            
-        return result