Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 341 files from 1 736 authors, of which 61 700 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
| NAP consistency checks, local citation audits, review response templates, local keywords. Uses firecrawl to scrape competitor GBP profiles and local SERPs.
| Revenue-focused keyword research and strategy for SEO campaigns. Handles keyword discovery, clustering, search intent mapping, difficulty assessment, competitor gap analysis, topical authority mapping, SERP feature analysis, competitor content strategy reverse engineering, and programmatic SEO template generation.
| 抖音视频解析、博主视频抓取/转录/运营分析/写入飞书。 触发词：解析抖音、抖音视频、抓取视频、博主视频、创作者视频、douyin、抖音解析、飞书多维表格 子命令：
Structure, classify, and write documentation using the Diátaxis framework. Use when writing docs, README files, guides, tutorials, how-to guides, API references, or organizing documentation architecture. Also use when asked to improve documentation, restructure docs, decide what type of doc to write, or classify existing content. Covers tutorials, how-to guides, reference, and explanation.
Use when reviewing code quality, asking whether a refactor or design is good, looking for code smells, or evaluating architecture tradeoffs. Identifies shallow interfaces, invalid state models, unclear boundaries, hidden effects, weak failure contracts, over-mocking, over-abstraction, compatibility glue, and clean-looking code that preserves bad models.
Use when reviewing PRs, simplifying over-engineered code, judging architecture, or renaming unclear concepts — inline fake helpers, delete meatless ceremony, rename lying names, merge over-split files, reject premature abstraction, say no to speculative config/modes/layers, then end with the smallest safe next change. Triggers when code is too fancy, too abstract, too clever, too many files/helpers/layers, or too well-factored but painful to change. Embody Grug brain: complexity very bad, small words, no consultant speak, no hard pivot to opposite dogma.
Use when creating or iterating staged planning artifacts for a complex feature: design discussion, structure outline, final executor plan, or supporting research/research questions. Also use when designing before coding, breaking work into vertical slices, or deepening a high-value roadmap item, NNN improve plan, architecture candidate, or HumanLayer-style task.
Use when building or reviewing frontend UI — dashboards, admin panels, landing pages, marketing sites, web apps. Drives domain-specific design decisions (typography, color world, layout, CSS token naming, depth and spacing systems) instead of generic AI defaults; routes to app.md (product/data UIs) or marketing.md (public/creative pages) by context.
Dude, come on. You write at the density of your own thinking and leave the reader to unpack it. Apply whenever explaining anything to a human — not just end-of-run summaries, every message.
Convert Docker Compose files to Coolify templates. Use when creating Coolify services, converting docker-compose.yml for Coolify deployment, working with SERVICE_URL/SERVICE_PASSWORD magic variables, or troubleshooting Coolify compose errors.
Use when the user asks to create, generate, visualize, or turn material into a standalone HTML artifact, self-contained or single-file .html report, interactive explainer, interactive illustration or visualization, visual walkthrough or handoff, or HTML presentation or slide deck. Handles source-grounded codebase and architecture, feature, PR or diff, implementation recap, incident timeline, option comparison, causal, and data-report explanations. Not for prose-only analysis or reviews, production web apps or dashboards, websites, HTML email, or ordinary frontend/UI implementation.
Use when answering complex questions about a codebase that require exploring multiple areas or understanding how components connect - coordinates parallel sub-agents to locate, analyze, and synthesize findings
Use when optimizing CLAUDE.md, AGENTS.md, custom commands, or skill files — diagnose the concrete failure first, then apply current documented Anthropic best practices (explicit instructions, context/motivation, examples, output and verbosity control, thinking/effort, CLAUDE.md size and skill-description rules) instead of inventing improvements. Triggers when a prompt isn't followed, a skill won't activate, CLAUDE.md is too long or ignored, or migrating prompts to current Claude models.
Use when evaluating designs, reviewing code, or refactoring - measures success by total code in the final codebase, not effort to get there. Bias toward deletion.
Survey a codebase as a senior advisor and turn the highest-value findings into implementation plans for other agents to execute — strictly read-only on source code, never implements anything itself. Use when asked to audit a codebase, find improvement opportunities (bugs, security, performance, test coverage, tech debt, architecture, migrations, DX), suggest features or roadmap direction, or generate handoff plans for another agent to implement.
Jujutsu (jj) — the Git-compatible version control system. Activate ONLY when a .jj/ directory is present in the project or when jj/jujutsu is explicitly mentioned. Do NOT activate for plain git repos without .jj/. Use for any VCS operations in jj-managed projects: commit, push, pull, branch, bookmark, rebase, squash, merge, diff, log, status, working copy, change ID, revset, fileset, template, configuration, workspaces.
Use when turning a roadmap, repo priorities, architecture priorities, chosen opportunities, accepted feature outlines/plans, or “roadmap to plans” request into grouped improvement plan batches for future executors. Writes README.md indexes, 001-*.md/NNN-style plans, optional memo-*.md files, verification gates, drift checks, STOP conditions, rejected approaches, and executor handoff notes.
Build evidence-backed software/product roadmaps from repo state, tickets, ideas, audit findings, or user goals — prioritize and sequence work across Now/Next/Later, milestones, dependencies, risks, and handoffs. Use when asked for a roadmap, feature roadmap, technical roadmap, release plan, milestone plan, backlog prioritization, sequencing, what to build next, or to turn improve/ideate/brainstorm/design-discussion outputs into an execution sequence. Not for implementing code or writing PR-level plans.
> Reviews Svelte components for idiomatic patterns, identifies anti-patterns carried over from React/Vue/Svelte 4, and suggests Svelte 5 refactorings. Use when writing or reviewing Svelte components, asking "is this idiomatic", "Svelte way", doing Svelte 5 migration, or slot-shaped APIs, lifecycle-driven code, imperative DOM wiring, immutable-update ceremony, context value replacement, shadcn-svelte form structure, Field.* components, bits-ui form controls, or component APIs that hide ownership. This is the general-purpose entry point for Svelte component review; delegates to sveltekit for routes/load/actions/server concerns and to focused references for details.
> Mental-model reset for Salsa, the incremental computation framework for Rust. Use when building or reviewing Salsa databases, tracked functions, input/ tracked/interned structs, query pipelines, accumulators, cancellation, durability, LSP integration, memory management, cycles, or production Salsa revisions, backdating, red-green algorithm, WillExecute, DidValidateMemoizedValue, Cancelled, returns(ref), no_eq, lru, cycle_fn,
> Mental-model reset for Rust. Use when writing or reviewing Rust code to shift from "it compiles" to "thinks in Rust." Triggers on Rust code review, "is this idiomatic", borrow-checker errors, API design, domain modeling, ownership, lifetimes, errors, traits, async/Tokio, unsafe, serde, FFI, tests, performance, Cargo structure, .rs files, Cargo.toml, rustc diagnostics, clippy findings, Result/Option, thiserror vs anyhow, newtype, typestate, enum vs trait, dyn Trait, Send/Sync, Pin, Miri, PyO3, napi-rs, cxx, UniFFI, wasm-bindgen, serde attributes, or feature unification.
Use when authoring, creating, refining, or troubleshooting agent skills — scaffold SKILL.md and frontmatter, write and optimize the trigger description, structure the body with progressive disclosure, validate structure, and test or debug activation. Also when building a new skill from scratch, when a skill won't trigger, loads incorrectly, or the agent ignores it entirely. Use when a skill misbehaved in the current session and needs adjustment based on learnings.
Use when a user asks “what should I work on in this repo?”, “what’s worth doing next?”, repo audit, tech-debt priorities, architecture priorities, old-plan reconciliation, or repo-grounded opportunity discovery. Produces .agents/ROADMAP.md with generated opportunities, user-seeded ideas, Now/Next/Later sequencing, rejected/not-now items, and next-artifact recommendations. Not for standard product/release roadmaps; if multiple repos are requested, create one roadmap per repo.
Use when writing, reviewing, or rewriting user-facing error messages, validation messages, form errors, empty/error states, auth errors, failure notifications, retry/support copy, CLI errors, or API errors humans will read. Helps make errors specific, actionable, non-blaming, accessible, and safe without leaking sensitive details.
Use when authoring an agent skill that wraps a command-line tool — covers hands-on tool exploration, required vs. recommended sections, installation/usage structure, trigger-rich descriptions, task-grouped commands, progressive disclosure, and a pre-publish checklist. Triggers for CLI / command-line / terminal / shell-command tools and binary wrappers; for review, run the Checklist section.
> Mental-model reset for SvelteKit apps. Use when writing or reviewing routes, layouts, load functions, form actions, remote functions, hooks, auth, cookies, endpoints, redirects, errors, SSR, progressive enhancement, or app-level data flow. Triggers on SvelteKit, +page, +layout, +server, +page.server.ts, +layout.server.ts, hooks.server.ts, load, actions, fail(), redirect(), error(), cookies, locals, route groups, protected routes, sessions, form actions, enhance, remote functions, command(), query(), form(), getRequestEvent(), SSR, hydration, and serialization. Use svelte5 for component-level runes, snippets, accessibility, actions, transitions, and component API review.
General prose doctrine for anything a human will read — clear, concrete, free of ready-made phrases. Use when writing or editing documentation, explanations, summaries, commit messages, reports, or any other prose.
Write self-contained implementation plan files that a fresh, possibly weaker executor can run without the author's context — splitting work into PR-sized plans, with verification gates, STOP conditions, and a memo protocol for design forks. Use when asked to write/create a plan, turn a decided design or ticket into plans, or when another skill hands off plan-writing. Not for deciding what to build or executing plans.
Generate images using Google Gemini with customizable options
Generate images using OpenAI gpt-image-2 with customizable options
Use when setting up guardrails for AI coding agents to enforce quality, security, and auditability requirements that must never be bypassed, or when configuring Claude Code lifecycle hooks in settings.json.
Use to create and maintain comprehensive documentation that helps agents navigate code effectively. Document directory structure, file purposes, architectural relationships, and design decisions so agents can work independently.
Use when starting a complex project, task, or feature to deeply clarify requirements, identify scope boundaries, map dependencies, and structure work into phases before any coding begins.
Use when working on multi-step tasks, long conversations, or when agent output quality degrades mid-session. Techniques for keeping context lean so the model produces better results.
Use when you need Claude to operate with domain expertise outside of software engineering — such as trademark law, go-to-market strategy, product management, security review, or any specialized field — and you want to write a short skill file that activates that expertise.
Use when starting a non-trivial coding task, feature, or bug fix that would benefit from upfront planning, research, and structured decomposition before writing code.
Use when creating new files, modifying existing files, adding new folders, or making structural changes to a codebase and you want documentation that agents (and humans) can grep/search to navigate the project.
Use when reviewing pull requests or critiquing code changes and you want high-signal, low-noise feedback by running multiple adversarial agents that challenge each other's findings. Works standalone or integrated into CI.
Use when setting up persistent memory for Claude agents across tools and projects, or when deciding where a piece of knowledge should be stored so the agent can recall it later.
Use when merging parallel agent work or branches to resolve both syntactic merge conflicts (merge markers) and semantic conflicts (contradictory logic, API changes, schema mismatches) that traditional merge tools miss.
Use when you need to research a codebase to inform implementation work — understanding call paths, data flow, schema shapes, or how existing features work — and you want facts without opinions leaking in.
Use when you need to research competitors, customers, market dynamics, or stakeholders and want Claude Code to synthesize findings into actionable product insights. Requires web access or pre-gathered research input.
Use when you need to run multiple Claude Code agents simultaneously on different parts of a codebase, or when a task is large enough to benefit from decomposition into parallel sub-tasks with clear contracts.
Use to comprehensively validate code quality and correctness in a separate context from the coding agent. Catches issues that biased validation would miss, ensures linting, tests, and assumptions are valid before merge.
Use when starting a new feature, writing a spec, or brainstorming architecture and you want to capture richer intent than typing allows. Speak your thoughts into a transcription tool, then feed the raw transcript to Claude Code for structuring.
Use when making user-facing UI changes and you want automated screenshot walkthroughs, or at release time to compare visual changes between releases and prioritize what needs human review. Requires Playwright.
Core conventions and patterns for this codebase
Use when the user wants to search Turkish academic journals on DergiPark (keyword or advanced field search by title/author/abstract/DOI/ORCID/year/etc.), read a DergiPark article PDF as text, or extract an article's references — drives the user's own Chrome (no CAPTCHA solving needed) by injecting JavaScript.
Use when troubleshooting, implementing, reviewing, or auditing platform infrastructure as a system — where Kubernetes, GitOps, CI/CD, and security concerns intersect. Provides structured diagnosis with blast radius, validation steps, and rollback plan for: Kubernetes, Flux CD, Argo CD, Terraform, GitHub Actions (composite actions, OIDC, SHA pinning), AWS, Azure, GKE, Linkerd, KEDA, Karpenter, supply chain security (Cosign, SBOM, SLSA), Falco, Chaos Engineering, DORA metrics, Datadog/Dynatrace/LLM observability, SOC 2, and PR review.
Commit changes with well-crafted messages, grouping related files into separate commits
Perform thorough code reviews focusing on unused code, duplications, coding patterns, bugs, and optimizations. Use when user wants code reviewed or audited. Read-only - outputs findings without making changes.
Capture current work context for handoff to another agent/developer. Gathers git state, todos, and modified files into a structured handoff document saved to the related spec folder.
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
Instrument web/web-app code with structured debug logging via a global variable (window.__debug_logs). Produces a clean JSON timeline for reproducing and diagnosing bugs. Use when user wants to debug a feature or track down a bug.
Create detailed implementation plans for features. Asks clarifying questions, suggests solutions, proposes architecture, and outputs a structured plan document. Use when user wants to plan a feature before coding.
Validate Claude Code skills against Anthropic guidelines. Use when user says "check skill", "skillcheck", "validate SKILL.md", or asks to find issues in skill definitions. Covers structural and semantic validation. Do NOT use for anti-slop detection, security scanning, token analysis, enterprise checks, or Eval Kit generation; use skill-check-pro for those. Do NOT use for LinkedIn skill engagement; use skillcheck-engage for that.
去除中文文本中的 AI 生成痕迹，使其更符合中国 MBA 毕业论文的自然写作风格。当用户提到"去 AI 痕迹"、"去除 AI 写作痕迹"、"MBA 论文改写"、"让这段文字更像人写的"、"去除机器感"等关键词时触发。适用于需要降低文本 AI 特征、提升学术写作自然度的场景。
Create a structured session handoff when context is running low or work is pausing. Deep context mining, self-validation, multi-file splitting. Captures everything the next session needs.
Run /handoff to capture session data, then write a phased implementation plan that references it. Creates beads for tracking.
Use when responding to a Revise & Resubmit decision from AER, AER:Insights, or an AEJ, and a point-by-point response letter plus aligned manuscript revisions are needed. Handles triage, the concede / clarify / push-back decision, and the response-letter format that editors actually read.
Claude Skills — questions
Answers built from the skills we actually parsed.