- Home
- Skills
- Image & Video
- Image Toolkit Studio
More screenshots
Works with the AI Agents you already use
Image Toolkit Studio
Resize, crop, convert and batch-export images. Preserve originals, control composition and transparency, and check every output.
Free
About this skill
Image Toolkit Studio is a complete free skill for preparing existing images with deliberate composition and verifiable output checks.
Start with your original files and intended destination. Choose from 80 practical task recipes across 20 scenario families, with 24 editable presets. Adjust them to your images and current destination requirements.
Useful everyday image preparation
Resize without stretching; fit a complete image onto an exact canvas; preview intentional crops with a chosen focus; rotate and normalize orientation; and export PNG, JPEG, WebP or AVIF where the installed codecs support them. Preserve transparency or explicitly choose a background matte. Create several output variants and batch-process images while keeping the originals unchanged.
The recipe library covers websites and blogs, email and newsletters, product listings, resale, social feeds, stories and mobile covers, profiles and teams, applications and upload limits, travel and family sharing, prints and postcards, logos and icons, developer handover, presentations, documents, scans and reference images, education, events, property and hospitality, food/pets/hobbies, and archives/privacy.
Checks you can inspect
Each build reports dimensions, format, file size, source fingerprints and unmet constraints. File-size targets are checked individually. When a target cannot be met within the requested dimensions and quality floor, the skill flags it instead of silently shrinking the dimensions or dropping below that floor.
Download individual outputs or an images-only ZIP. A separate full review pack includes reports, a portable plan and uncropped source previews; review its contents before external sharing. Saved plans contain settings and fingerprints, so restoration requires the exact originals. Metadata removal does not anonymize visible faces, text or other image content.
Flexible execution
The helper can execute directly from loaded text resources in a Python interpreter. An optional local workbench provides uploads, previews, batch variants, per-image adjustments, undo/redo and saved-plan recovery when the environment permits a saved local workspace. The browser workbench is optional and is not promised in an interpreter-only host.
Runtime: Python 3.10+ and Pillow 12.3+ with ImageCms/LittleCMS. WebP and AVIF depend on installed codecs. HEIC/HEIF needs a separately available optional decoder. No paid API, seller backend, telemetry or image upload to an external service is required by the image helper. Software is not installed automatically.
What is included and what has been verified
The 19-file English text core contains instructions, helper code, local-workbench resources and recipe data. Original code and written resources are MIT licensed; image rights and third-party software licenses remain separate.
The gallery shows real local outputs from fictional, synthetic demonstration photos and authored technical fixtures. Demo photographs and generated example binaries are not included in the text core. Local validation rebuilt all 80 demonstrations, yielding 169 image outputs with no encoder failures; five outputs honestly report unmet byte targets. Representative images were visually reviewed; not every output received individual subjective review.
Agensi delivery disclosure: auxiliary-resource loading, buyer-side execution and generated-file retrieval have not yet been verified. Local interpreter and workbench validation does not establish marketplace delivery compatibility. Generated user images and reports are distinct from protected Skill source and must use output storage and retrieval allowed by the host.
This skill prepares still images. It does not provide AI retouching, background removal, restoration, OCR, vector editing, animation export, identity-photo certification or professional press preflight. Presets do not guarantee current platform acceptance or certified print color.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.
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
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 26 days ago
- Free with an Agensi account
- Software
Python