11 atomic image processing skills covering background removal, enhancement, OCR, ID photos and more. Zero dependencies, install and use instantly — give your AI Agent professional-grade image processing capabilities.
Each skill maps to one PicWish API endpoint, carefully wrapped and ready to use
AI-powered foreground detection for people, objects, and stamps. Precisely removes backgrounds and outputs transparent PNGs with auto or manual subject type selection.
E-commerce · Design · AssetsPortrait-optimized cutout with refined edge handling for hair strands, ears, and fine details. Outputs transparent background headshots.
Avatars · Badges · ProfilesAI super-resolution to boost clarity and resolution. Supports general enhancement and face-specific mode (up to 10 faces). Auto or manual 1x/2x/4x scaling.
Photo Restore · Print ReadyMask-based intelligent object erasing. Mark the area to remove and AI seamlessly fills it with surrounding content, like content-aware fill.
Remove People · Erase TextFully automatic watermark detection and removal — no mask needed. AI identifies text and logo watermarks and cleanly fills the area.
Asset Cleanup · Batch ProcessingOne-click standard ID photo generation. Auto-crops to proper head-shoulder ratio, supports passport, visa, and various national formats with custom background colors.
Passport · Visa · ApplicationAI deep learning colorization for grayscale photos. Intelligently infers colors based on semantic context — sky turns blue, grass turns green, skin gets natural tones.
Photo Restore · HistoricalSmart compression that minimizes file size while preserving visual quality. Compress by quality level or target file size, with optional resize.
Web Optimization · StorageHigh-accuracy OCR engine to extract text from images. Supports multi-language recognition and outputs to txt, pdf, docx, xlsx, or pptx formats.
Digitization · Invoice ExtractAuto-detects document or object edges for perspective correction and intelligent cropping. Supports document enhancement and object background removal.
Doc Scan · Card CorrectionPixel-level semantic segmentation separating clothing, body parts, and accessories into independent masks. Multiple granularity levels from clothing-only to full body.
E-commerce · Virtual Try-on · Fashion AIMore image processing skills are on the way. Stay tuned…
Four steps to give your AI Agent image processing capabilities
Install all 11 image processing skills via ClawHub with a single command.
clawhub install picwish-skills
Set up your PicWish API Key via either method below. Visit https://picwish.com/my-account to get a free API Key.
# Option A: Environment variable
export PICWISH_API_KEY="your_api_key_here"
# Option B: OpenClaw config
openclaw config set skills.entries.picwish.apiKey "your_api_key_here"
If your API key is from picwish.cn, please set the region to cn. Overseas users can skip this step (defaults to global).
# Option A: Environment variable
export PICWISH_REGION=cn
# Option B: OpenClaw config
openclaw config set skills.entries.picwish.region cn
The AI Agent automatically routes user intents to the right skill.