Docs & Writing
Docs & writing agent skills turn AI agents into dependable technical writers: READMEs, API references, PDFs and Word documents, changelogs, and style-guide enforcement. Install one and your agent produces the same structure and tone every time it writes.
-
superclaude-org Skill TroubleshootSystematic troubleshooting with root cause analysis. Use when users report errors, bugs, or unexpected behavior. Never retry without understanding why.
-
akash-network Bundle CLI DesignUse when building, improving, or reviewing command-line interfaces for better user experience - before implementing commands/output/errors, when users report confusion or frustration, or when CLI feels hard to use - provides UX principles, visual design techniques, and practical patterns for creating discoverable, delightful CLIs
-
citability Skill StandingReport where the user actually stands, from evidence. Reads the last N Claude Code session transcripts plus git, extracts the work that shipped and the invisible work that did not leave a diff, compares it against a trailing baseline from the user's own history (and an industry source only when a cited one exists in the repo), and states objective progress with every number tied to an instrument and a time window. Use when the user asks "how am I doing", "am I making progress", "where do I stand", "was this week worth it", or is comparing themselves to others in their field. Never invents a benchmark; says Unknown when it cannot measure.
-
dddfxyqiming Bundle Video Notes GeneratorSummarize video URLs or local videos into timestamped Markdown notes from complete transcripts and actual visual evidence, with adaptive sampling and targeted visual follow-up. Supports Bilibili, YouTube, Douyin and Kuaishou.
-
erickramirezau Bundle Youtube Video EvaluatorEvaluates a batch of YouTube video candidates (typically the output of the youtube-idea-finder skill) for thumbnail quality, editing quality, topics, tech/tools, and format, then turns each into a Replicate/Fix/Calendar-action brief. Use this after youtube-idea-finder has surfaced overperforming videos, whenever the user wants to know WHY a video worked so they can copy the parts that worked and avoid the parts that didn't, or asks to turn found video ideas into a content calendar. Does not search or filter videos itself — it takes a list of already-identified candidates and produces a downloadable Markdown diagnostic report.
-
gongzitang Bundle Mpvault ArchiveTurn all accessible historical articles from a WeChat public account into a resumable, locally imaged, searchable, verified Markdown knowledge vault. Use when Codex is asked to 导出公众号、抓取微信公众号历史文章、把公众号转成 Markdown、建立公众号知识库、搜索已归档文章、续传公众号下载、核验公众号归档完整性, or operate MPVault with a target original ID such as gh_xxx.
-
gongzitang Bundle Build Persona VaultBuild and maintain one verified, cross-platform knowledge vault for a public person or creator from WeChat public-account and Weibo archives, with full-first then incremental synchronization, stable source IDs, provenance, unavailable-state evidence, Markdown/JSONL exports, and verification. Use when Codex, Claude, or WorkBuddy is asked to 建立人物知识库、汇总某个人的公众号和微博、跨平台归档创作者内容、持续增量更新人物语料、合并 MPVault 与微博 Markdown, or operate PersonaVault.
-
hybb-rash Bundle Readable Report把已有报告重组为更容易读懂的阅读版,保留理解结论所需的依据与边界,按需加入图解和交互。用于报告可读性改造、长报告重排或可视化阅读版;不用于从零调研、核验事实或单独索要几句摘要。
-
hahaknight Skill Ppt OutlinePPT outline & slide-by-slide content design (PPT 大纲与逐页内容设计)。输入主题和受众,输出完整的逐页 PPT 内容稿(可直接粘贴进 PPT 或 WPS)。当用户说"帮我做个PPT大纲"、"明天汇报的PPT"时使用。(数字员工配置包免费样例;完整版 6 岗位技能+自动防护+3套SOP 仅 ¥19.9,购买/免费评审拷贝见 github.com/Hahaknight/digital-employee-pack)
-
hahaknight Skill Weekly ReportWeekly report generator from git commits & work logs (周报生成器)。把本周的提交记录、待办完成情况、聊天中提到的进展,整理成一份结构清晰的周报。当用户说"写周报"、"生成周报"、"这周干了啥"时使用。(数字员工配置包免费样例;完整版 6 岗位技能+自动防护+3套SOP 仅 ¥19.9,购买/免费评审拷贝见 github.com/Hahaknight/digital-employee-pack)
-
hugoveltorai Bundle Tufte VizDesign, build, and critique data visualizations using Edward Tufte's principles. Use this skill WHENEVER the work involves a chart, graph, plot, dashboard, table, diagram, or any visual display of data, even if the user does not name Tufte or "data viz" explicitly. Triggers include: (1) Building a new chart, graph, plot, or figure (bar, line, scatter, time-series, small multiples, sparkline, etc.) (2) Designing or reviewing a dashboard, report, or slide with quantitative visuals (3) Critiquing or improving an existing visualization (4) Choosing between visualization approaches ("what's the best way to show this?") (5) Reducing chartjunk, improving the data-ink ratio, or fixing a misleading chart (6) Planning high-density displays, comparisons, or part-to-whole views Applies data-ink ratio, chartjunk elimination, graphical integrity, lie factor, small multiples, layering, sparklines, and analytical design. Consult before producing any chart so output is clear, honest, and high-density.
-
jschobl Bundle Run TestsRun the pytest suite, report pass/fail counts and coverage, and identify untested code. Use when the user asks to run tests, check test coverage, or verify that changes didn't break anything.
-
jschobl Skill Format PythonFormat Python source files with Black and isort, then report any remaining lint issues. Use when the user asks to format, clean up, tidy, or fix the style of Python code.
-
jayofemi Bundle WordingLint user-visible text before it ships, catching em and en dashes, a colon used in place of an em dash, AI-tell phrasing, the word "copy" misused for site text, and jargon or surfaced internal concerns like funding on non-technical surfaces.
-
ziwang-physics Bundle Agentchat Independenttasks并行分发 N 道独立任务给多个浏览器 AI,回收后由 Claude Code 合成权威答案并输出教科书式解答手册 PDF。触发词:"分给多个 AI""并行问多个 AI""汇总成 PDF/解答手册"。不适用于需 AI 间协作或角色分工的任务。
-
zzxx-bit Bundle Openclaw Deep ResearchPlan and execute deep research, literature reviews, due diligence, competitive analysis, policy comparison, market mapping, or investigative reports in OpenClaw with source-verifiable claims, local ledgers, and optional delegation. Use when the user asks for a high-quality report, wants citations or auditable evidence, wants research process files written locally, or needs help managing context-window limits in an OpenClaw workflow.
-
zzxx-bit Bundle Opencode Deep ResearchPlan and execute deep research, literature reviews, due diligence, competitive analysis, policy comparison, market mapping, or investigative reports in OpenCode with source-verifiable claims, local ledgers, and optional delegation. Use when the user asks for a high-quality report, wants citations or auditable evidence, wants research process files written locally, or needs help managing context-window limits in an OpenCode session.
-
zzxx-bit Bundle Claude Code Deep ResearchPlan and execute deep research, literature reviews, due diligence, competitive analysis, policy comparison, market mapping, or investigative reports in Claude Code with source-verifiable claims, local ledgers, and optional delegation. Use when the user asks for a high-quality report, wants citations or auditable evidence, wants research process files written locally, or needs help managing context-window limits in a Claude Code workflow.
-
5zjk5 Bundle Project Analyzer代码仓库梳理与项目文档生成技能。当用户明确要求梳理项目或使用项目梳理功能时触发此技能。典型触发场景包括"帮我梳理这个项目"、"梳理一下这个代码仓"、"使用项目梳理"等。此技能读取项目代码仓,按照结构化模板生成完整的项目梳理 Markdown 文档。
-
david-saeteros Bundle Paper CompanionStructured academic paper reading guide using a four-stage protocol: Structural Scan → Core Findings → Critical Evaluation → Reading Note. Use this skill whenever a user uploads a PDF of an academic paper, or says things like "let's read this paper", "help me with this article", "walk me through this study", "Paper Companion", or "stage 1/2/3/4". Also trigger when the user asks to analyse, summarise, or critically evaluate any research paper — even without a direct upload. Do NOT use the academic-writing skill in place of this one for paper-reading tasks.
-
kenliu0607 Bundle Notion2docxThis skill can be used when users need to convert Notion pages to Microsoft Word (.docx) files. This skill preserves formatting, colors, and tables, and converts LaTeX formulas to editable OMML Word formulas.
-
kenliu0607 Bundle Ppt CompressCompress PPT/PPTX file size. Decompress PPT, compress large images, repackage and convert to PDF to significantly reduce file size. Suitable for scenarios where large PPT files need to be shared or uploaded.
-
nazmussayad Skill DocsGuides documentation for the right audience. Use when creating, updating, reviewing, or restructuring documentation, README files, guides, tutorials, or references.
-
opencoworkai Bundle PDFComprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.
-
opencoworkai Bundle DOCXComprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks
-
opencoworkai Bundle PPTXPresentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks
-
chrislacey89 Skill CounselConsult the local book library for frameworks relevant to a task or decision, using metadata routing (each book's category, tags, and situational index) rather than blind full-text search. Use when the user says '/counsel <question>', 'consult the library', 'what do my books say about X', 'check the library before drafting', 'which framework applies here', or when another skill calls for a library counsel step. Output is a short counsel report: 2-4 frameworks, each cited to book-slug/references/file.md and applied to the task at hand. Read-only against the library.
-
davideast Bundle Diataxis DocumentationGuides the creation, restructuring, and review of technical documentation using the Diátaxis framework. Use when writing docs, auditing existing docs for "smells", or structuring complex documentation hierarchies.
-
elliothux Bundle Cf Compatibility CheckReview branch and working-tree implementation changes for conformance with open-compute's Cloudflare Workers runtime target under explicit single-machine self-host exclusions. Use when checking new Workers, Durable Objects, Queues, Workflows, R2, D1, KV, binding, type, or runtime behavior against current Cloudflare; report evidence-backed findings without editing. Do not use for Cloudflare management API parity or general code review.
-
finos Bundle Technical WriterAssists with writing and maintaining Morphir technical documentation. Use when creating, reviewing, or updating documentation including API docs, user guides, tutorials, and content for the Docusaurus site. Also helps ensure documentation quality through link checking, structure validation, and code review for documentation coverage.
-
jayminwest Skill Researching External SourcesUse when researching documentation, papers, or articles for insights. Triggers on "research external", "find sources", "literature review", "what do the docs say".
-
yucong-28 Bundle Markdown Math WriterUse when creating, editing, organizing, completing, or rewriting Markdown/.md files — READMEs, course notes, research notes, math/physics documents, formula derivations, lab reports — to write LaTeX math correctly per TARGET renderer. Target routing: README/GitHub/repository docs → github-safe mode: inline single-dollar math, math-fence blocks, equation numbers as Markdown text, no tag command, no aligned, no boldsymbol; Markdown Preview Enhanced notes → mpe mode: double-dollar blocks, aligned and tag allowed; otherwise generic. Triggers: 创建Markdown/修改.md/README/GitHub README公式/学习笔记/科研笔记/数学文档/公式推导/LaTeX公式修复/Markdown Preview Enhanced文档/公式竖排错乱/Extra close brace.
-
zohaibcodez Bundle Skill MakerCreates new Claude Code skills following official standards and best practices. Use when you need to generate a new skill, create skill documentation, or scaffold a skill structure. Automatically validates naming, descriptions, and ensures quality standards.
-
a2aproject Skill Release A2aGuide maintainers through the multi-step release process — version bump, CI verification, tagging, Maven Central deployment, SNAPSHOT bump, and versioned documentation.
-
aboudjem Skill Sniff ReportShow the results from the last sniff scan.
Audited -
adewale Bundle Slide MakerGenerate presentation decks grounded in real GitHub projects, or walk through a structured brief-to-slides process. Use when the user asks to create a presentation, slide deck, talk, pitch, keynote, or Slidev project — especially when they want slides based on an existing codebase, architecture, or project documentation.
Frequently asked questions
What are Docs & Writing agent skills?
Docs & writing agent skills turn AI agents into dependable technical writers: READMEs, API references, PDFs and Word documents, changelogs, and style-guide enforcement. Install one and your agent produces the same structure and tone every time it writes.
Which Docs & Writing skills are most installed?
Popular Docs & Writing skills on SkillMD right now include troubleshoot, cli-design, standing. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do Docs & Writing skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.