add feature allow user to paste image

Author: daltonnyx

AgentCrew/__init__.py

@@ -1 +1 @@
-__version__ = "0.6.5-4"
+__version__ = "0.6.6"

AgentCrew/modules/clipboard/service.py

@@ -1,13 +1,22 @@
 import pyperclip
 import base64
 import io
+import os
+import tempfile
 from PIL import ImageGrab, Image
-from typing import Dict, Any
+from typing import Dict, Any, Optional
+import logging
+
+logger = logging.getLogger(__name__)
 
 
 class ClipboardService:
     """Service for interacting with the system clipboard."""
 
+    def __init__(self):
+        """Initialize the clipboard service."""
+        self.temp_files = []  # Keep track of temporary files for cleanup
+
     def write_text(self, content: str) -> Dict[str, Any]:
         """
         Write text content to the clipboard.
@@ -30,6 +39,36 @@ def write_text(self, content: str) -> Dict[str, Any]:
                 "error": f"Failed to write to clipboard: {str(e)}",
             }
 
+    def _create_temp_file_from_image(self, image: Image.Image) -> Optional[str]:
+        """
+        Create a temporary file from a PIL Image.
+
+        Args:
+            image: PIL Image object
+
+        Returns:
+            Path to the temporary file or None if failed
+        """
+        try:
+            # Create a temporary file
+            temp_fd, temp_path = tempfile.mkstemp(
+                suffix=".png", prefix="clipboard_image_"
+            )
+            os.close(temp_fd)  # Close the file descriptor
+
+            # Save the image to the temporary file
+            image.save(temp_path, format="PNG")
+
+            # Keep track of temp file for cleanup
+            self.temp_files.append(temp_path)
+
+            logger.info(f"Created temporary image file: {temp_path}")
+            return temp_path
+
+        except Exception as e:
+            logger.error(f"Failed to create temporary file from image: {str(e)}")
+            return None
+
     def read(self) -> Dict[str, Any]:
         """
         Read content from the clipboard and automatically determine the content type.
@@ -72,6 +111,71 @@ def read(self) -> Dict[str, Any]:
                 "error": f"Failed to read from clipboard: {str(e)}",
             }
 
+    def read_and_process_paste(self) -> Dict[str, Any]:
+        """
+        Read clipboard content and if it's an image or binary file, create a temporary file
+        and return a file command that can be processed.
+
+        Returns:
+            Dict containing either processed file command or regular text content
+        """
+
+        image = ImageGrab.grabclipboard()
+
+        clipboard_result = {
+            "success": False,
+        }
+
+        if image is not None and isinstance(image, Image.Image):
+            # Handle image content - create temporary file
+            temp_file_path = self._create_temp_file_from_image(image)
+
+            if temp_file_path:
+                clipboard_result = {
+                    "success": True,
+                    "content": temp_file_path,
+                    "type": "image_file",
+                    "format": "file",
+                    "cleanup_required": True,
+                }
+
+        if not clipboard_result["success"]:
+            return clipboard_result
+
+        content_type = clipboard_result.get("type")
+
+        if content_type in ["image_file", "binary_file"]:
+            # Return a file command for the temporary file
+            temp_file_path = clipboard_result["content"]
+            return {
+                "success": True,
+                "content": f"/file {temp_file_path}",
+                "type": "file_command",
+                "temp_file_path": temp_file_path,
+                "original_type": content_type,
+                "cleanup_required": clipboard_result.get("cleanup_required", False),
+            }
+        elif content_type == "image":
+            # For base64 images, still return as is for backwards compatibility
+            return clipboard_result
+        else:
+            # Regular text content
+            return clipboard_result
+
+    def cleanup_temp_files(self):
+        """Clean up any temporary files created by this service."""
+        for temp_file in self.temp_files:
+            try:
+                if os.path.exists(temp_file):
+                    os.unlink(temp_file)
+                    logger.info(f"Cleaned up temporary file: {temp_file}")
+            except Exception as e:
+                logger.warning(
+                    f"Failed to cleanup temporary file {temp_file}: {str(e)}"
+                )
+
+        self.temp_files = []
+
     def write(self, content: str) -> Dict[str, Any]:
         """
         Write content to the clipboard.
@@ -83,3 +187,7 @@ def write(self, content: str) -> Dict[str, Any]:
             Dict containing success status and any error information
         """
         return self.write_text(content)
+
+    def __del__(self):
+        """Cleanup temporary files when the service is destroyed."""
+        self.cleanup_temp_files()

AgentCrew/modules/console/input_handler.py

@@ -18,6 +18,7 @@
 
 from AgentCrew.modules import logger
 from AgentCrew.modules.chat import MessageHandler
+from AgentCrew.modules.clipboard.service import ClipboardService
 from .completers import ChatCompleter
 from .display_handlers import DisplayHandlers
 from .constants import (
@@ -41,6 +42,7 @@ def __init__(
         self.console = console
         self.message_handler = message_handler
         self.display_handlers = display_handlers
+        self.clipboard_service = ClipboardService()
 
         # Threading for user input
         self._input_queue = queue.Queue()
@@ -75,6 +77,40 @@ def _(event):
             # This will be handled by the main console UI
             pass
 
+        @kb.add(Keys.ControlV)
+        def _(event):
+            """Handle Ctrl+V with image/binary detection."""
+            try:
+                # Check if clipboard contains image or binary content
+                paste_result = self.clipboard_service.read_and_process_paste()
+
+                if paste_result["success"]:
+                    content_type = paste_result.get("type")
+
+                    if content_type == "file_command":
+                        # Insert the file command
+                        file_command = paste_result["content"]
+
+                        # Insert the file command into the current buffer
+                        event.current_buffer.insert_text(file_command)
+                        event.current_buffer.validate_and_handle()
+
+                        return
+
+                # For regular text content, use default paste behavior
+                event.current_buffer.paste_clipboard_data(
+                    event.app.clipboard.get_data()
+                )
+
+            except Exception:
+                # Fall back to default paste behavior if anything goes wrong
+                try:
+                    event.current_buffer.paste_clipboard_data(
+                        event.app.clipboard.get_data()
+                    )
+                except Exception:
+                    pass  # Ignore if even default paste fails
+
         @kb.add(Keys.ControlC)
         def _(event):
             """Handle Ctrl+C with confirmation for exit."""
@@ -257,7 +293,7 @@ def get_user_input(self):
                 style=RICH_STYLE_BLUE,
             )
             title.append(
-                "\n(Press Enter for new line, Ctrl+S/Alt+Enter to submit, Up/Down for history)\n",
+                "\n(Press Enter for new line, Ctrl+S/Alt+Enter to submit, Up/Down for history, Ctrl+V to paste)\n",
                 style=RICH_STYLE_YELLOW,
             )
             self.console.print(title)

AgentCrew/modules/gui/components/input_components.py

@@ -3,16 +3,16 @@
 from PySide6.QtWidgets import (
     QVBoxLayout,
     QHBoxLayout,
-    QTextEdit,
     QPushButton,
     QCompleter,
     QFileDialog,
     QSizePolicy,
 )
-from PySide6.QtCore import Qt, QStringListModel
+from PySide6.QtCore import Qt, QStringListModel, Slot
 from PySide6.QtGui import QTextCursor
 import qtawesome as qta
 from AgentCrew.modules.console.completers import DirectoryListingCompleter
+from AgentCrew.modules.gui.widgets.paste_aware_textedit import PasteAwareTextEdit
 from .completers import GuiChatCompleter
 
 
@@ -29,8 +29,9 @@ def __init__(self, chat_window):
 
     def _setup_input_area(self):
         """Set up the input area with text input and buttons."""
-        # Input area
-        self.chat_window.message_input = QTextEdit()
+        # Input area - use our custom paste-aware text edit
+        self.chat_window.message_input = PasteAwareTextEdit()
+
         input_font = self.chat_window.message_input.font()
         input_font.setPixelSize(16)
         self.chat_window.message_input.setFont(input_font)
@@ -46,6 +47,7 @@ def _setup_input_area(self):
         self.chat_window.message_input.setSizePolicy(
             QSizePolicy.Policy.Minimum, QSizePolicy.Policy.MinimumExpanding
         )
+        self.chat_window.message_input.image_inserted.connect(self.image_inserted)
 
         # Create buttons layout
         buttons_layout = QVBoxLayout()
@@ -75,6 +77,10 @@ def _setup_input_area(self):
         # Store the buttons layout for use in main window
         self.buttons_layout = buttons_layout
 
+    @Slot(str)
+    def image_inserted(self, file_command):
+        self.chat_window.llm_worker.process_request.emit(file_command)
+
     def _setup_file_completion(self):
         """Set up file path completion for the input field."""
         # Set up file path completion

AgentCrew/modules/gui/widgets/paste_aware_textedit.py

@@ -0,0 +1,52 @@
+from PySide6.QtWidgets import (
+    QTextEdit,
+)
+from PySide6.QtCore import Signal
+from AgentCrew.modules.clipboard.service import ClipboardService
+
+
+class PasteAwareTextEdit(QTextEdit):
+    """Custom QTextEdit that handles paste events to detect images and binary content."""
+
+    image_inserted = Signal(str)
+
+    def __init__(self, parent=None):
+        super().__init__(parent)
+        self.clipboard_service = ClipboardService()
+
+    def insertFromMimeData(self, source):
+        """Override paste behavior to detect and handle images/binary content."""
+        try:
+            # Check clipboard content using our service
+            paste_result = self.clipboard_service.read_and_process_paste()
+
+            if paste_result["success"]:
+                content_type = paste_result.get("type")
+
+                if content_type == "file_command":
+                    # It's an image or binary file - use the file command
+                    file_command = paste_result["content"]
+
+                    self.image_inserted.emit(file_command)
+
+                    # Show status message
+
+                    return  # Don't call parent method
+
+                elif content_type == "text":
+                    # Regular text content - let the parent handle it normally
+                    super().insertFromMimeData(source)
+                    return
+
+                else:
+                    # Other content types (like base64 image) - handle normally for now
+                    super().insertFromMimeData(source)
+                    return
+            else:
+                # Failed to read clipboard, fall back to default behavior
+                super().insertFromMimeData(source)
+
+        except Exception as e:
+            # If anything goes wrong, fall back to default paste behavior
+            print(f"Error in paste handling: {e}")
+            super().insertFromMimeData(source)

README.md

@@ -116,6 +116,7 @@ Here are some things AgentCrew can do:
   window (GUI).
 - **File Handling:** AI agents can work with text and image files in chat.
   AgentCrew also supports PDF, DOCX, XLSX, and PPTX files.
+- **📋 Smart Paste Detection:** Automatically detects images and binary content when pasted (Ctrl+V). Images from screenshots, copied files, or other sources are automatically converted to `/file` commands for seamless processing.
 - **Streaming Responses:** Get real-time replies from AI agents.
 - **"Thinking Mode":** Some AI models can show their reasoning process.
 - **Rollback Messages:** Easily go back to an earlier point in your

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "agentcrew-ai"
-version = "0.6.5-4"
+version = "0.6.6"
 requires-python = ">=3.12"
 classifiers = [
     "Programming Language :: Python :: 3",