Crop and prepare provided source images for Ylang Labs blog artwork, especially `cardImage.png` and `blogHeader.png`, for posts in the ylang-labs-blog project. Use this skill whenever the user provides an image and asks to make blog images, crop blog artwork, create a card image, create a blog header, resize a header, fit an image to blog requirements, or update article thumbnails. The agent must visually analyze the provided picture first and choose the best crop for each required aspect ratio instead of blindly center-cropping.
Add Ylang Labs black or white logo watermarks to blog source artwork and derived cover assets. Use when a user asks to watermark Ylang Labs blog images, apply the Ylang Labs logo mark, choose a black or white logo by local contrast, or prepare source, header, and card assets with the brand mark.
Create Ylang Labs blog cover packages that combine an original museum-quality oil painting with an exact official technology logo. Use this skill whenever a user asks for a source artwork, card image, blog header, cover image, social post image, or blog artwork that should place a real model, framework, product, or company logo over a painterly background. Generate the painting first, fetch the official logo separately, composite it onto the square source image, derive the crops, add the Ylang watermark, and validate every final asset.
Create beautiful, repeatable image-generation prompts and raster section visuals for Ylang Labs technical blog posts. Use this skill when the user asks for inline section images, technical diagrams, exploded views, blueprint diagrams, architecture plates, process maps, system diagrams, product-teardown visuals, spec sheets, or OpenAI image-generation prompts in the clean editorial technical style of the provided reference images. For normal Ylang Labs blog cover artwork, `cardImage.png`, and `blogHeader.png`, use `oil-painting-image-generator` first, then `blog-image-cropper`; use this skill for cover/card/header assets only when the user explicitly asks for a technical-diagram cover. Always pair generation with OpenAI's `imagegen` skill and the latest official GPT Image / ChatGPT image generation path; do not create SVG/vector/code-native diagram sources as part of this skill unless the user explicitly asks for manual vector work.
Review, critique, fact-check, and validate Ylang Labs blog drafts or published posts with a factuality-first standard. Use this skill when the user asks to review a blog, check factual accuracy, assess MDX/frontmatter/assets/citations, find unsupported claims, flag overlong or overcomplicated drafts, review simplicity, clarity, and concision, or provide editorial feedback before publishing. Use `blog-writing-guide` as the primary editorial rubric for Ylang Labs blog prose.
Create and structure Ylang Labs blog MDX posts, including slug selection, frontmatter, asset directories, reference packets, and MDX component conventions. Use this skill when the user wants to create a new blog post file, start a Ylang Labs article, set up blog assets, or convert a content brief into repo-ready MDX. Use `blog-writing-guide` as the prose standard and `content-calendar-management` when the post comes from or should create a calendar issue.
Coordinate the complete Ylang Labs blog publishing workflow from idea or draft through writing guidance, MDX setup, optional artwork or technical visuals, cropped card/header images, social copy, validation, draft PR publication, and optional content-calendar tracking. Use this skill whenever the user asks to create, publish, prepare, or turn a blog idea/draft into a full Ylang Labs blog post, especially when they want writing, assets, social posts, scheduling, GitHub issues/projects, or a PR handled together.
Default Ylang Labs blog creation and drafting skill. Use this skill first when the user asks to create, write, draft, ideate, or turn an idea into a new Ylang Labs blog post, especially when no final direction has been selected yet. Orchestrates a five-subagent blog variation workflow by having subagents produce five distinct article variants with cover and inline image specs, then coordinates selection or synthesis into a final MDX-ready draft. The main agent assigns repo blog writing, MDX, image, cropper, factuality review, cleanup, and publishing skills to subagents rather than doing that specialist work itself. Also use when the user explicitly asks for multiple blog variations, five or more subagents/agents, parallel writing agents, or a main agent to coordinate different writing styles and image directions. If the user only asks to package, prepare, or publish an already selected final draft, use blog-publishing-workflow instead.