topographic
Topographic terrain with the topographic library: 2D contour maps and 3D landscapes, generated or from real elevation data, for websites, videos and games. Use…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-09 00:13
Design & creation
22,585 results
pentest-task-design
Overall guidance for penetration testing security testing task design and execution. As the top-level orchestration skill for penetration testing, it maps abstract security testing methodologies to directly callable toolchains, defining the complete decision process from target intake to report output. It covers every specific operation in reconnaissance, crawling, vulnerability testing, and exploitation verification, linking recon-planning, web-crawler, xss-testing, …
style-transfer
Based on the ArtShift gallery number or the user's photo, generate style transfer prompts and images; when no number is selected, recommend temporary original directions.
dongfang-cover-design
Design landscape Editorial covers, vertical posters, or square communication graphics using Eastern aesthetics as the method, and either output a visual plan with image-generation prompts or directly generate finished images. Suitable for content covers, social media square images, people and ideas, cities and solar terms, cultural events, museum exhibitions, art publishing, brand communication, e-commerce promotions, and commercial advertising; first use visual direction to match the subject and communication goal.
handraw-browser-producer
When the browser output route has been selected, produce hand-drawn videos and picture-book videos; execute the seven storyboard types, programmatic animation, unified compositing, and MP4 acceptance. Share the plan and asset thresholds with the video master control.
handraw-jianying-producer
Native draft and asset delivery package for making hand-drawn videos and picture-book videos when the CapCut output route has been selected; by default limited to ST-001–004 with item-by-item checks of native effects, automatic web pre-rendering prohibited.
handraw-video-producer
Hand-drawn video master control: produce hand-drawn videos, animated picture books, and narrated picture books. First choose browser final output or Jianying draft, confirm the plan and review materials, then dispatch the corresponding production Skill. Shares seven types of storyboards and external video continuations.
knowledge-video-director
Design the goals, evidence, scripts, and storyboards for knowledge and tutorial videos; implement only using the seven storyboard categories, confirm the plan first, then collect materials and produce.
mascot-ip-studio
Coordinate original mascot requests for projects, accounts, or folders, directly producing three distinct complete IP candidates; exclude personal-photo asset…
rongbao-illustrations
Generate Chinese body illustrations that can select a registered IP by name (default: Yazai; includes Rongbao, Abao, and confirmed registered personal characters); when the user simultaneously mentions a character, this IP, explicitly invokes $rongbao-illustrations or carries an IP, as well as cover, vertical poster, square image, mengli/character anchor/turnaround/3:4 infographic/sticker, Baoyu/宝玉's content capabilities, or real-person photo personal...
design-dna
Translate a vague visual direction, reference website URL, named product, or optional screenshot into an original, executable design language and apply it to a…
saas-illustration
SaaS product hero illustration style — single brand accent color + real product UI card + line-art scene. For website hero sections, landing pages, feature sections, blog headers, presentation title slides, and adjacent scenarios (tech talk covers, portfolio featured projects, workshops…
cli-anything-photoshop
CLI harness for controlling Adobe Photoshop via COM automation. Supports project management, layer operations, selection control, text editing, image adjustment, and multi-format export.
video-clipper
Batch-generate short-video clips from long videos (livestream replays, meeting recordings, podcasts). Precisely locate opinion boundaries based on transcripts, automatically remove silent pauses, stutters and disfluencies, and output short videos with synchronized audio and video. Suitable for: livestream clipping, meeting highlight extraction, short-video secondary editing, podcast highlight clips.
cli-gpt-image
Generate or edit images from the terminal using the user's LOCAL ChatGPT/Codex login (no API key, no proxy). Use when asked to create, generate, draw, render,…
brand-guidelines
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand…
canvas-design
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art,…
image-gen
Invoke: /img, $img Generate images from text descriptions using ZETATECHS API. Use whenever the user wants to create, generate, or make images, illustrations,…
social-card
Use when the user asks to generate a social media card image, OG image, Twitter card, link preview, or social share graphic. Do NOT use for full-page…
design2code
Read .design/ specs → generate real pages and components in the target project
chatgpt-image-generator
Use when the user asks Codex to open ChatGPT in the in-app browser, generate one or more images with their own ChatGPT account, export originals locally, batch…
breathing-brand
Minimalist breathing-feel visual full package for going-global / international brands. Use when the user says they want brand VI, brand visuals, logo and merchandise, independent-site visuals, or brand identity for a community / organization / product, or complains that their brand "looks cheap," "too full and cluttered," or "not international / high-end enough," or wants brand design with "whitespace," "breathing room," or a "high-end feel." Produce positioning insight, color + whitespace system, line-link motif, lines...
pixel-sprite
Convert the subject of a reference image or text description into a Low Resolution Pixel Sprite (low-resolution pixel illustration / pixel icon). When the user uploads an image and says "make it pixel style / pixel art / pixel art / 8-bit / 16-bit / sprite / game icon / pixel avatar", or wants to turn...
taste-pm
Use when designing a product's structure, information architecture, page layout, entry points, or user-experience flow — turning a business or feature need…