ideacco/baoyu-skills-openclaw
15 skills15 installs180 starsGitHub
Install
npx skills add https://github.com/ideacco/baoyu-skills-openclawSkills in this repo
1Baoyu Article IllustratorBaoyu Article Illustrator is a Generative Media agent skill for solo and indie builders who publish articles and need on-brand visuals without opening a separate design tool. You feed it article or section content; it applies fixed infographic rules—hand-drawn only, horizontal 16:9, readable hand-lettered text—and invokes nano banana pro to render the illustration. A gated first-time setup collects preferences in EXTEND.md when that file is missing, so repeat runs stay consistent. It fits the Grow phase, especially content workflows where posts, tutorials, and newsletters need scannable graphics that reinforce keywords and core ideas. Use it when you already have copy and want a single cohesive infographic frame, not a photo mockup or UI screenshot. The skill optimizes for editorial clarity and agent-driven batching alongside your writing stack rather than general-purpose image playgrounds.1installs2Baoyu Compress ImageBaoyu Compress Image is a Bun-driven compression skill for solo builders who ship marketing sites, SaaS dashboards, or storefronts with heavy screenshots and hero art. The agent runs a small CLI that walks folders (optionally recursive), picks the best available toolchain on the host—Apple sips, Google cwebp, ImageMagick, or sharp—and writes smaller outputs while reporting compression ratios. Use it when Lighthouse flags oversized images, when you need WebP variants without hand-tuning every asset, or when you want JSON summaries for automated PR checks. It stays phase-specific to shipping performance work rather than creative direction: you still choose quality and format, but the skill removes repetitive ffmpeg-style trial and error across machines.1installs3Baoyu Cover ImageBaoyu Cover Image is a generative-media agent skill that chooses cover art direction from what you are publishing, not from a fixed template. When type, palette, or rendering inputs are left blank, tables map content cues—product launches, architecture essays, quotes, narratives, zen minimal pieces—to hero, conceptual, typography, metaphor, scene, or minimal layouts, then pair them with warm, elegant, cool, dark, earth, vivid, pastel, mono, or retro palettes and flat-vector, hand-drawn, painterly, digital, or retro digital renderings. Solo builders shipping newsletters, WeChat posts, SaaS launch graphics, or blog headers can invoke it during Launch distribution or Grow content workflows so every piece gets a deliberate visual lane without hiring a brief writer. It complements copy-first skills by closing the thumbnail gap before posts go out.1installs4Baoyu Danger Gemini Webbaoyu-danger-gemini-web is an agent skill packaged as a TypeScript integration layer around Gemini’s web-facing RPC patterns—cookie rotation, access tokens, candidate parsing, and image-generation helpers. Solo builders who already automate Claude Code or Codex tasks but need Gemini images or web-model access without standing up a full Google Cloud project install it during the build phase when wiring LLM backends. The readme emphasizes exception taxonomy, timeout fetch wrappers, and gem-oriented types so agents fail predictably instead of returning opaque HTML errors. Treat it as high-risk integration: unofficial web endpoints can break, rate-limit, or violate terms of service, so you should pair usage with your own guardrails and never commit live session cookies to git. It does not replace official Gemini API SDK setup for production SaaS; it is for builders experimenting with web-session automation inside personal or internal tooling.1installs5Baoyu Danger X To Markdownbaoyu-danger-x-to-markdown (first-time setup) is a blocking onboarding skill in the baoyu OpenClaw family. When EXTEND.md is missing, the agent must not start converting tweets or articles, ask for URLs, or pick output paths until the user answers bundled preference questions and the file is written. Questions cover how to handle images and videos—prompt each run, always download locally, or keep remote links—and the default output directory for markdown exports. Language follows the user’s input or saved preference. After EXTEND.md exists, the main conversion workflow can proceed with stable defaults. Solo builders use this to avoid inconsistent archives when scraping X content into a notes or publishing pipeline. It is narrow and procedural: zero conversion logic here, only gating setup. Skip if EXTEND.md is already present and still accurate.1installs6Baoyu Format Markdownbaoyu-format-markdown is an agent skill for solo builders who ship a lot of Markdown—READMEs, changelogs, skill packages, and bilingual content—and want repeatable cleanup instead of one-off editor fixes. It encodes a Node/remark stack (GFM, frontmatter, CJK-friendly rules) plus an optional autocorrect step so agent output matches publishing conventions before you merge or distribute. Use it in the Build docs lane whenever drafts look fine in chat but break lists, spacing, or frontmatter in the repo. It does not replace content strategy or SEO; it makes technical writing mechanically correct and easier to review in PRs.1installs7Baoyu Image GenBaoyu Image Gen (first-time setup flow) is an agent skill for solo builders who want repeatable AI image generation inside an OpenClaw/Baoyu-style skill pack without re-answering provider questions every session. When no EXTEND.md exists, the skill walks through default provider, model, and preferences in one structured interview; when EXTEND.md exists but default_model for the chosen provider is null, it runs a shorter model-selection-only path. That persistence layer matters for indie workflows that batch covers, UI mocks, and social creatives from the same agent. Use it during Build when you integrate multimodal APIs into content or agent tooling, before any batch image job. The skill is configuration and workflow guidance—it does not replace API keys, billing, or your host app's asset pipeline. After setup completes, continuation flows assume EXTEND.md is the contract for downstream baoyu-image-gen generation steps in the same repository.1installs8Baoyu InfographicBaoyu Infographic is an agent skill that generates professional infographics from markdown files, URLs, or pasted text. It treats layout (how information is structured) and style (how it looks) as independent choices, with 21 layouts and 20 styles you can combine freely. The skill analyzes your content, suggests sensible layout×style combinations, and produces publication-ready visuals suited to tutorials, product explainers, and dense summaries. Solo builders shipping newsletters, landing pages, or internal docs can invoke it when they need a visual summary instead of another wall of text. Defaults are configurable via flags for layout, style, aspect ratio, and locale (including zh and ja). It is a generator workflow migrated from the baoyu-skills lineage, not a live charting API integration.1installs9Baoyu Markdown To Htmlbaoyu-markdown-to-html is an agent skill that compiles markdown into standalone HTML using a typed style system, theme CSS, and an image pipeline that downloads remote assets and rewrites placeholders. Solo builders use it when a draft already lives in markdown—changelog, landing copy, tutorial, or newsletter—and they need a polished HTML file without hand-tuning templates in a CMS. The flow covers renderer initialization, code-theme CSS, structural HTML modifications such as removing duplicate headings, and optional backups keyed by timestamps so iterations stay reversible. It complements URL-to-markdown when your pipeline is capture → edit → publish. Complexity sits at intermediate because theme paths, image hosting, and meta fields must be coherent for acceptable output on first run.1installs10Baoyu Post To Wechatbaoyu-post-to-wechat is an agent skill for posting markdown articles to WeChat Official Accounts. It documents a bun-based script that converts markdown (or HTML) into editor-ready content, supports multiple visual themes, and handles the awkward WeChat image workflow by inserting placeholders during render and swapping real images in the browser editor afterward. Solo builders who write in markdown but need a Chinese social distribution channel can invoke the skill so their agent runs the right npx command, respects frontmatter for title and author, and explains parameter choices. It fits builders doing content-led growth rather than building a WeChat mini-program codebase.1installs11Baoyu Post To Xbaoyu-post-to-x is an agent skill backed by a Bun script that opens Google Chrome and loads your Markdown into the X Articles editor with formatting and images. Solo and indie builders who write essays, launch threads as articles, or repurpose blog Markdown for X can stay in one source file instead of reformatting by hand. Frontmatter and the first H1 supply the title; cover and inline images are resolved locally or downloaded to a temp folder before placeholder replacement in the UI. Preview mode is the default so you can review layout before `--submit` actually publishes. You need an X Premium subscription for Articles, Chrome installed, and Bun available for the script runner. It fits the launch phase when distribution is the bottleneck, not when you are still validating copy or building the product itself.1installs12Baoyu Slide DeckBaoyu slide deck is an agent skill that converts source material into professional presentation assets: an outline with style instructions, individual slide images, and merges into PowerPoint or PDF via bundled TypeScript scripts. Solo and indie builders use it when asked to create slides, make a presentation, generate a deck, or produce PPT-style output without opening a separate design tool chain manually. The workflow supports presets, custom styles, audience tuning, multilingual output, and outline-only passes so you can approve structure before image generation. Placement spans Validate when you need pitch and scope clarity, Launch when you prepare distribution narratives, and Grow when you repurpose content into webinar or social funnels. Allowed tools include read, write, and bash so agents can run merge scripts from the skill directory. Complexity is intermediate because you must point agents at SKILL_DIR scripts correctly and choose style parameters that match your brand. The skill is a generator package migrated from the baoyu/openclaw ecosystem, not a live slide SaaS integration.1installs13Baoyu Url To Markdownbaoyu-url-to-markdown is an agent skill that turns live web pages into markdown files using Chrome and the DevTools Protocol. Solo builders use it when they need trustworthy text from JS-heavy sites—competitor landings, API docs, blog posts—without hand-copying or brittle curl scrapes. The implementation centers on a CDP WebSocket client, profile-aware browser launch, and fetch helpers with explicit timeouts so agents do not hang on network idle. It fits early journey work (research and audience discovery) and later reuse when you refresh docs or repurpose articles. Pair it with markdown-to-HTML or summarization skills when the goal is a publishable artifact rather than a raw capture. Expect filesystem and browser permissions because the skill drives a real browser profile on your machine.1installs14Baoyu Xhs Imagesbaoyu-xhs-images is an agent skill package for producing Xiaohongshu-ready image sets with a mandatory first-run preference capture. Before the main creative pipeline runs, the agent detects a missing EXTEND.md and executes a blocking setup: watermark choice, default style (or custom name), and persistence to EXTEND.md—without asking about articles, audiences, or layouts prematurely. Solo and indie builders who post on XHS use it to standardize generated visuals (handles, branding, style defaults) so later automation stays consistent. It fits builders already using Claude Code or similar agents inside the baoyu-skills-openclaw workflow and who want repeatable social creatives rather than one-off prompts. The skill is workflow-oriented: setup completes, then the skill hands off to Step 1 of the broader XHS image pipeline described in the repo.1installs15Release SkillsRelease Skills is a universal agent workflow for solo and indie builders who ship libraries, CLIs, SaaS backends, Claude plugins, or generic repos and want one command instead of five stack-specific playbooks. It reads project metadata, picks the right version file, applies semver bumps you choose or infer, updates changelogs, and drives the git-oriented release steps agents can run with read, write, and bash. The skill fits the Ship phase when you are ready to tag and push—not when you are still scoping features. It complements testing and review skills by assuming artifacts are merge-ready. Dry-run mode makes it safe to rehearse releases on side projects or monorepo packages before you announce a version publicly.1installs