image delivery
Batch process, resize, and SKU-map images on Windows with transactional undo and CSV verification reports.
Free
Works with the AI Agents you already use
image delivery
Example session with this skill installed
Process the images in C:\Photos\Summer2024 using the web-optimized-webp preset and the sku-map.json. Run a dry run first to check for collisions.
- Read your context and instructions
- Compiled the image delivery
- Generated the image
I have performed a dry run for the image processing task. The system scanned the source directory and validated the SKU mappings against the web-optimized-webp preset.
Dry Run Results
| File Name | SKU | Target Format | Status | Action |
|---|---|---|---|---|
| DSC001.jpg | SKU-8821 | WebP | Valid | Resize & Convert |
| DSC002.png | SKU-8822 | WebP | Valid | Resize & Convert |
| DSC003.jpg | SKU-8823 | WebP | Valid | Resize & Convert |
| logo_v1.webp | SKU-9001 | WebP | Valid | Format Pass-through |
Validation Summary:
- Source Files: 4
- Potential Collisions: 0
- Format Errors: 0
- Validation: All files successfully mapped to SKUs.
The dry run confirms that the output directory is empty and all 4 images meet the input criteria. No files were written during this step.
Next steps
- Execute the actual batch by running the process command with
dry_run: false. - Retrieve the
operation_idfrom the successful batch to run a verification report. - Use
verify_outputto generate the final CSV compliance manifest.
DSC001.jpg
PNG · 1536×1024
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Preparing hundreds of product images for an e-commerce storefront or catalog usually involves a messy mix of manual resizing, naming errors, and failed byte-size constraints. This skill provides a specialized MCP server interface to automate batch image processing with strict schema validation and transactional safety.
What it does
- Batch Processing Automates resizing, format conversion, and padding for JPEG, PNG, and static WebP files.
- SKU Naming Maps source filenames to specific SKU identifiers using JSON mapping schemas.
- Compliance Verification Runs post-process audits to verify SHA-256 hashes, pixel decodability, and byte-size limits.
- Atomic Undos Reverts the most recent batch operation on Windows, cleaning up output files while preserving unrelated data.
- Validation Reporting Generates CSV acceptance reports for internal QA or client delivery.
How it works
- Inventory Source Use
list_imagesto scan the local source directory and identify valid image assets. - Configure Preset Select a JSON preset for dimensions and format, then provide a SKU map for final naming.
- Dry Run Execute
process_imageswithdry_run: trueto catch naming collisions or mapping errors without writing files. - Execute & Verify Run the actual batch to generate outputs, then use
verify_outputwith the operation ID to confirm spec compliance. - Rollback Call
undo_lastif the batch needs to be reverted due to errors or changing requirements.
Frameworks & tools
Built with .NET 10 and SkiaSharp for high-performance image manipulation on Windows x64 systems. It utilizes the Model Context Protocol (MCP) to bridge local file operations with AI agents.
Why this beats prompting it yourself
Standard LLM prompting cannot interact with your local file system or perform precise pixel-level resizing and hashing. This skill adds a dedicated C# engine that handles the heavy lifting of image encoding while providing the agent with structured feedback on every file processed.
Use cases
- Preparing thousands of product photos for Shopify or Amazon via SKU mapping.
- Converting high-res photography into web-optimized WebP batches with strict byte limits.
- Validating large image deliveries against a specific technical manifest before ingestion.
- Standardizing disparate folder structures into a unified, named asset library.
Known limitations
Supports Windows x64 only. Reparse points, network paths, and alpha-preserving outputs are not supported. Only the most recent active batch can be undone.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.
Requires Windows x64 and .NET 10 SDK. No support for network paths or transparent backgrounds. Batch undo is limited to the most recent operation.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 6 days ago
- Free with an Agensi account
- Software
- .NET 10