Shape agent capabilities into your workflow.

This Claude Code skill fetches public WeChat articles from the user's machine, searches by topic, reads selected results, and caches repeated requests locally. It is designed for public `mp.weixin.qq.com/s/...` share URLs and does not cover articles restricted to followers or other non-public pages.

This Claude Code skill stores boards, linked issue specifications, and cross-session context inside a project-local `kanbanboards/` folder. The files remain readable in any editor with no build step or runtime dependency, while the optional Obsidian Kanban plugin adds an interactive drag-and-drop view.

This Claude Code collection uses a router-to-executor architecture: project-compass identifies the current stage and directs work to a state-machine SOP builder, ACTOR reading coach, Socratic discussion partner, skill auditor, or postmortem writer. Teams can install the full set or reuse the routing pattern with their own executors.

This Claude Code skill turns a natural-language motion concept into a live HTML playground for iterating on timing, typography, and transitions. The final animation uses plain CSS and JavaScript and can be exported as a React component, with ten reusable motion patterns and a complete worked example included for reference.

OSS Community Health helps maintainers interpret the human side of an open source project, including contributor retention, onboarding friction, concentrated workload, bus-factor risk, and burnout patterns. It favors multi-week trends and pairs each observation with a concrete next step rather than acting as another metrics dashboard. Discord, Slack, or forum signals are optional and require explicit consent, community notice, and aggregate-only analysis without individual attribution.

Run a morning, afternoon, or evening check-in and let the plugin turn the conversation into a first-person Obsidian note. It can identify an emotional “floor” on a 34-level scale, track selected habits, and add an optional advisory panel below a clear divider so commentary is not mixed with the user’s own voice. Claude Code and an Obsidian vault are required; the anonymous first-load install ping can be disabled.

expert-debate turns an idea into two deliberately conflicting professional proposals instead of a balanced pros-and-cons list. Claude writes from one expert role, then produces a copyable prompt for an independent Model B session with the opposing role and no shared context. You compare the arguments and record your reasoning manually, so the workflow needs no API keys or automated model integration.

blind-review prepares a prompt for a fresh Model B session that evaluates a document without knowing the author’s intent or prior collaboration context. Comparing that independent reading with what you meant can reveal missing assumptions and “knowledge curse” gaps before a PRD or specification reaches stakeholders. An optional devil’s-advocate pass adds pressure when reviewers agree too quickly, and the workflow uses manual copy-paste rather than an API.

doc-generator converts a finalized product idea into development-ready documentation, including a PRD, technical specification, CLAUDE.md, and implementation guide. Its PRD flow includes an AI-simulated user-validation step, and the command supports a full four-document suite, PRD-only output, a unified PRD plus blueprint, or an interactive from-scratch mode. This lets teams and solo builders choose the amount of structure they need.

Run a weekly digest over local Claude Code JSONL sessions to measure work patterns such as activity mix, one-shot edit rate, agent usage, model allocation, and project attention. When a diagnostic threshold fires, the plugin appends a concrete rule to CLAUDE.md or adds an investigation item, then measures the same behavior again in later reports. It also sends a one-time anonymous install signal unless the documented opt-out environment variable is set.

Shape an academic manuscript around the conventions that apply to its paper type and intended venue, rather than applying one generic writing style. The skill asks for those two inputs, then guides section structure, argument moves, voice, citation format, LaTeX choices, and reviewer-risk checks across six paper types and five venue categories. It does not invent citations, search the literature, verify facts, create figures, or run statistical analysis.

Give the skill a market topic and optional business context, and it gathers material from nine source types through Claude Code’s web tools and the GitHub CLI. It extracts signals, clusters themes, builds a Playing to Win strategy with inline citations and falsifiable assumptions, then adds objective and expert-style scores. The result is a self-contained HTML report with linked evidence, while saved stage files make an interrupted run resumable.

Give it a research question and the Claude Code skill decomposes it, searches the open web with Tavily, and, when relevant, adds GitHub, academic indexes, or current library documentation. It grades and corroborates sources before producing a research plan, cited report, source file, and claim-to-evidence map. Claude Code, Tavily MCP, and Python 3.10 or later are required.

webcpu-agent-skill gives coding agents verified references, examples, and templates for building browser-based Canvas2D scenes with EASEL.js 0.5.0. It packages the same guidance for Claude, Codex, OpenCode, Copilot, and other agent environments, with a rule to report unknown APIs rather than invent them. Rendering still requires a browser DOM canvas; Deno can type-check, bundle, and serve the app but does not provide headless Canvas2D by itself.

Clawproof Checks evaluates an agent system across ten operational areas, including least privilege, audit trails, prompt injection, human escalation, rollback, secrets, and regression testing. The packaged Claude skill can inspect a codebase, runbook, or configuration and return a scored report with prioritized remediation. The same checks are also available as YAML and consolidated JSON for vendor-neutral programmatic use.

routine-writer teaches agents to express inbox triage, pull-request reviews, alert checks, briefings, trackers, and fixers as terse routines rather than long operational playbooks. It defines a title, tagline, and body pattern with verb-first instructions, explicit iteration, optional actions, empty-state handling, and a short report. The skill follows the Agent Skills specification and can be installed in compatible harnesses.

Fix My Resume interviews you about your experience, target role, and real achievements before building or rewriting an ATS-friendly single-column DOCX. It diagnoses vague bullets, passive voice, repeated verbs, formatting problems, and missing job-description keywords, then provides a heuristic score and tuning workflow. It refuses to invent metrics and asks for evidence instead; language checks are English-focused, and neither the score nor the resume can guarantee interviews.

html-to-pdf-agent-skill uses Playwright to render remote HTML pages as local PDFs. It can handle one page or wait for a single-page application to load, extract real documentation links from the rendered sidebar, convert each page, and merge the results. Page size, margins, styles, and authentication can be customized; Playwright with Chromium is required, and full-site merging additionally depends on PyPDF2.