[FEATURE]

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

When I drag an image into the Claude Code terminal, the image content is received and visible to Claude, but
the file path is not provided. This creates a disconnect: Claude can see and analyze the image but doesn't
know where the file is located. This is particularly useful when I want Claude to not only analyze an image
but also help me work with the file itself (move it, copy it, rename it, etc.).

Proposed Solution

When a user drags an image file into the terminal, Claude Code should include both the image data (so Claude
can see and analyze it) and the file path (so Claude knows the file location). This way, if Claude analyzes
an image and I ask it to do something with the file, it already has the path available and can help
immediately.

Alternative Solutions

_No response_

Priority

Critical - Blocking my work

Feature Category

CLI commands and flags

Use Case Example

Use Case Example: Image Upscaling with Custom Skill

When I drag a photo into the Claude Code terminal, I want to ask Claude to upscale it using my custom skill
that integrates with the fal.ai image upscaling API.

My skill workflow requires:

1. File path of the dragged image (e.g., /path/to/image.jpg)
2. Read the image file using fs.readFileSync(filePath)
3. Convert the image to base64 encoding
4. Send a POST request to: https://api.fal.run/fal-ai/image-upscaler
5. Include the base64-encoded image in the request body with parameters like scale_factor: 2
6. Receive the upscaled image from fal.ai API response
7. Save the processed image back to the original directory

Current workflow (requires manual intervention):

1. Drag image → Claude sees image but NOT file path
2. User: "Upscale this image"
3. User must manually provide the file path
4. Skill can now process the image

Desired workflow (with this feature):

1. Drag image → Claude receives BOTH image data AND file path automatically
2. User: "Upscale this image"
3. Skill immediately has the path, reads the file, and sends to fal.ai API endpoint
4. No manual path entry needed

This eliminates the extra step and allows the skill to access the file path directly from the drag-and-drop
event, enabling seamless integration with external APIs.

Additional Context

_No response_