Engineering
98,276 rows, by stacks then stars
emaddawood1991-tech/claude-ai-skills/skills/mcp-web-creative-toolkit Skill
no license0★ repoCoordinate Playwright, Firecrawl, Clipia, and Chrome MCP for browser QA, web research, crawling, and AI media generation. Use for MCP setup, tool selection, and safe multi-tool workflows.
Preetam3620/chatbot-creator-skill Skill
no license0★Scaffold a standalone LLM chatbot app (FastAPI + React + Vite). Builds the full project — user auth, multi-turn chat with Claude/OpenAI/Gemini, WebSocket streaming, SQLite persistence. Warns before overwriting existing files.
jedobe/skill-evaluator/skills/skill-evaluator Skill
0★ repoScore a Claude Code skill against a research-backed rubric derived from the top 9 most-starred skill repos on GitHub (obra/superpowers 229k★, affaan-m/ECC 216k★, anthropics/skills 151k★, and more). Use when a user asks to evaluate, score, review, rate, or improve a skill — or when they share a SKILL.md and want to know how good it is. Do NOT use for CLAUDE.md, .cursorrules, or system prompt files — these are not packaged skills; note the file type in the output and score what you can.
JustKurayy/rd-council/.claude/skills/rd-council Skill
no license0★ repoRun an R&D planning meeting on this codebase. Dispatches three Haiku "seats" (Senior Engineer, Security Engineer, Product/UX) to analyze the code in parallel, has them cross-critique each other once, then synthesizes a ranked roadmap of what to build, fix, or improve next. Use when the user invokes /rd-council, or asks "what should we work on / build / improve next", wants a code health + roadmap review, or wants a multi-perspective audit of next steps.
0xdenny218/vercel-geist-design-skill/skills/vercel-geist-design Skill
0★ repoApply Vercel's Geist design system to frontend and app UI work. Use when designing, building, refactoring, or reviewing web apps, Next.js pages, dashboards, developer tools, docs sites, product surfaces, and agent-generated UI that should feel Vercel-like: minimal, high-contrast, restrained, token-driven, accessible, and developer-focused.
Rishiidev/handoff-pro Skill
0★Eliminates starting-from-zero when context fills up or a session ends. Two modes — SAVE compresses the current session into a structured handoff document; LOAD reads a previous handoff and resumes with full context as if the conversation never ended. Invoke immediately when the user says any of: "handoff", "create handoff", "save handoff", "load handoff", "load latest handoff", "resume", "load session", "save session", "pick up where we left off", "continue from last session", or when the context window is visibly filling up. In Cowork and Claude Code, loading is fully automatic — no copy-paste needed. Works across Claude.ai, Cowork, and Claude Code.
Tingtao-code/skill-intake Skill
0★Vet a candidate skill / MCP server / prompt before adopting it into your Claude Code setup, then optionally install it. Checks overlap with skills you already have, fit to your actual needs, weight (is it bloated?), distillation depth, and — for MCP/prompts — tool-permission & data-exfil safety. Ends in a verdict (ADOPT / DISTILL-THEN-ADOPT / REWRITE / REJECT) and one-shot landing. USE FOR: "evaluate this skill", "should I install this MCP", "vet this prompt", pasting a third-party SKILL.md / MCP config, deciding whether a new capability earns a place in your library. SKIP FOR: authoring a skill from scratch (use a skill-authoring skill), routine code/debug/design work. See also: write-a-skill (downstream when the verdict is REWRITE).
rifatshampod/vibeops/codex/skills/vibeops-assess Skill
0★ repoAssess a project's infrastructure needs and recommend how to deploy it. Use this whenever the user wants to ship, deploy, host, publish, or 'put online' an app, or asks how to get their project live — even if they don't say the word 'infrastructure'. Also triggers when a user shares a project and asks 'what do I need', 'where should I host this', 'how much will it cost', or 'is my app ready to deploy'. Read-only and safe to run anytime.
tdrapp/agentic-architecture Skill
0★Best practices for DESIGNING and STRUCTURING Claude agent systems — anchored on Anthropic's 'Building Effective Agents' taxonomy (workflows vs agents; augmented LLM; routing, parallelization, orchestrator-workers, evaluator-optimizer) and Claude Code / Agent SDK guidance, extended with validated cross-project patterns. Use this whenever architecting a multi-agent or subagent system, deciding workflow-vs-autonomous-agent, structuring skills/subagents/tools, wiring context isolation, adding a verification/red-team pass, encoding lessons as reusable memory, or packaging a Claude Code workspace as an Agent SDK app (dual-mode). Trigger even if the user only says 'how should I structure this agent', 'should this be a subagent or a tool', 'how do I split this into agents', 'design the orchestration', or 'turn my Claude Code repo into a real app'. NOT for generic application code (use toti-engineering) or a specific product stack (use genai-saas).
ndisisnd/kermit/skills/kermit Skill
0★ repoFormat and run git commits using Conventional Commits style with emoji prefix, point-form file bodies, and BREAKING CHANGE footer. Manages CHANGELOG.md via hooks. Use --init to initialize changelog.
nmelanitis/reviewer-skill Skill
no license0★Use this academic citation-review skill when the user wants to analyze how later papers cite or use a target paper. It supports OpenAlex citation expansion from a DOI or OpenAlex work ID, regex or internal-agent/cloud-LLM classification of citing papers as yes/no/dont_know, optional open-PDF download and full-text extraction, review-prompt construction, and cautious literature-review drafting. Trigger for literature-review, citation-review, paper-usage review, related-work synthesis, OpenAlex citation analysis, and "which papers use this paper" tasks. Do not use for generic software code review.
alectivism/organization-ai-skills/org-agents/skills/subagent-delegation Skill
0★ repoUse BEFORE starting any multi-part task (research or comparisons across several sources, reading or summarizing long documents and transcripts, bulk updates across many items, audits, or checking a draft or numbers before they go out). Also use when about to spawn any subagent or worker, or when a task needs 3+ searches or tool calls. Routes gathering, reading, and bulk work to cheaper worker models, and escalates hard reasoning or a final check to a stronger model only when that actually helps.
waseemnasir2k26/skynetlabs-all-claude-code/skills/client-delivery-and-planning/client-onboard Skill
no license0★ repoOne command bootstraps a per-client repo so Claude Code starts pre-loaded. Detects project type (WP theme / React-Next / static / n8n / video) from files present → generates a tight CLAUDE.md (<200 lines: stack, architecture rationale, conventions + build/deploy commands) with the RIGHT deploy gotcha baked in automatically (Hostinger Node app = manual ZIP NOT GitHub; React = Vercel; video = NVENC-broken→libx264). Scaffolds .claude/, suggests relevant MCP (GHL/Hostinger), adds CLAUDE.local.md to .gitignore, enforces the CLAUDE.md-vs-Skill rule, and warns on lowercase claude.md dups. Fixes the gap: many client repos (travel, agency site, jewelry, fashion, car dealer, health clinic) ship with no CLAUDE.md, so Claude re-learns context every session. Trigger when user says: "onboard this client", "onboard repo", "bootstrap CLAUDE.md", "set up CLAUDE.md", "init client repo", "/client-onboard", or opens a client repo with no CLAUDE.md and asks Claude to start work on it. Auto-trigger heuristic: cwd is a client project dir AND no CLAUDE.md exists at repo root.
LaoitdevOpen/laoitdev-skills/skills/frontend/feature-architecture Skill
0★ repoFeature folder structure and architecture for React + TypeScript admin apps. Use this when creating a new feature, adding a major section, or when asked how to organize a feature's files. Covers the canonical folder layout (components, hooks, services, types, schemas, constants, utils), what goes where, and index export conventions.
Salesforce architect navigator
ayanme/salesforce-navigator-suite/skills/salesforce-architect-navigator Skill
0★ repoThis Navigator researches Salesforce architecture patterns, integration design, and enterprise guidance from official architect documentation. Use when the user asks how to design or evaluate Salesforce architecture: integration patterns, reference architectures, scalability, security design, governance, identity strategy, enterprise trade-offs, Agentforce architecture, Data Cloud architecture, Marketing Cloud architecture including data extensions, send architecture, subscriber data models, and Marketing Cloud Personalization, CRM Analytics architecture, MuleSoft, and API-led connectivity.
SkillMedev/data-analyst-toolkit/skills/ab-test-analyzer Skill
0★ repoAnalyzes an A/B experiment to a defensible verdict - sample-size and minimum-detectable-effect math, two-proportion significance testing with confidence intervals, and checks for the traps that fake a win (peeking, sample ratio mismatch, multiple comparisons). Use when someone asks "is this test significant", "did variant B win", "how long should I run this experiment", "what sample size do I need", or shows experiment results and asks whether to ship. Do NOT use for logging and organizing ML training runs - use experiment-tracking instead; for causal questions without a randomized experiment use causal-inference; for general funnel or metric investigation use funnel-analysis or product-analytics.
The same file is in 2 other repositories
takeaseatventure/cloudcost-doctor Skill
0★Audit infrastructure-as-code (Terraform, CloudFormation, CDK) for cloud cost anti-patterns BEFORE deployment. Finds overprovisioned instances, expensive storage defaults, idle resources, missing lifecycle policies, and 45+ more — no AWS credentials required. Use when reviewing or cost-optimizing IaC.
siam-hossain9/secure-development-skill/plugins/secure-development/skills/secure-development Skill
0★ repoSecurity best practices for building and reviewing software, web apps, APIs, and infrastructure. Use whenever writing, designing, modifying, or auditing code — especially anything that handles user input, authentication, authorization, sessions, secrets, cryptography, databases, file uploads, network/HTTP requests, APIs, LLM/AI features, realtime/WebSockets, payments, PII, deployment/cloud/containers, or when reusing/pasting code (including AI-generated or copied Stack Overflow snippets) — to prevent vulnerabilities (injection, XSS, SSRF, CSRF, IDOR/broken access control, auth flaws, secret leakage, prompt injection, misconfiguration) by design rather than patching them later. Also use for security reviews, threat modeling, and pre-ship hardening.
KijitoAI/kijito-claude/skills/kijito-qa-memory Skill
0★ repoRigorous Kijito memory curation with enforced creation + cold-boot verification. Use when winding down a session, before /clear or self-clear, when asked to "QA / curate / clean up memory", when preparing a handoff for the next session, or any time you need to be sure a fresh session could continue the work. Counters the two chronic failure modes — treating "QA" as corrections-only (skipping creation), and never confirming the memory actually works in a cold context.
justhandledlabs/skills/plugins/env-doctor-free/skills/env-doctor-free Skill
0★ repoDiagnose local project environment issues that prevent apps from starting or running. Use when the user says "why won't this run", "check my environment", "env doctor", "diagnose startup issue", "it works on my machine", or asks for help debugging missing dependencies, runtime versions, port conflicts, .env problems, file permissions, or stopped services.
The same file is in 2 other repositories
eUmeda/LLM-prompting-bestpractice Skill
0★外部 LLM サービス (Anthropic Claude, OpenAI GPT, Edison Scientific Kosmos, Elicit, Perplexity 等) に投入するプロンプトを 3 ベンダー統合 best practice で設計・最適化する。XML タグ構造化・5 コンポーネント (role/task/context/format/constraints)・outcome-first 指示・Kosmos 用 single-objective 設計・Anti-pattern 検査を整合的に適用する。Triggers on: プロンプト最適化, prompt optimization, prompt engineering, Kosmos プロンプト, Edison プロンプト, GPT プロンプト, Claude プロンプト, XML タグ, prompt 設計, prompt review, 外部 LLM 投入, /LLM-prompting-bestpractice
maverickgao8848/researcher-agent Skill
0★2 Agent 并行深度调研系统。当用户需要深度调研某个话题、生成专业扎实的调研分析文章时激活。 触发词:深度调研、研究报告、research、调研报告、专题研究。 架构:Claude(协调+写作)→ 2 个 Research Agent(并行搜索)→ 综合写作 → 可选 Mermaid 配图。 输出:实用主义视角的调研分析文章,只说有用的关键内容,输出到 output/{topic}/research/article.md。
edilkoke/vibeguard Skill
0★Use when building, reviewing, or hardening a web application with AI assistance — especially "vibe-coded" apps on Next.js + Supabase. VibeGuard enforces secure-by-default code generation and reviews output against a catalogue of the 21 most common AI-coding vulnerabilities (exposed secrets, broken access control, injection, weak auth, missing privacy controls, vulnerable dependencies). Trigger on: "build an app", "add auth", "is this secure", "review my code", "harden", "deploy", or any handling of payments, identity, logins, or customer data.
nishimoto265/anthropic-dev-loop/skills/anthropic-dev-loop Skill
0★ repoUse when the user wants to run, document, or reuse an Anthropic-inspired Claude Code development loop, including planning interviews, Fixture/Invariant/Probe verification, main/sub-agent implementation, PR review evidence, and post-merge harness updates. Trigger on phrases like Anthropic DevLoop, Claude Code development loop, harness-driven development, evals, verification loop, or multi-agent implementation flow.
SSBrouhard/npm-package-release-hardening Skill
0★Use when releasing or publishing a JavaScript/TypeScript npm package or CLI publicly — making a package public-ready, reviewing package.json or tarball contents, setting up GitHub CI, Dependabot, security settings, or branch protection, triaging dependency-update PRs, or when an agent-built package may overstate the Node runtime it supports.
EthanQC/SystemWright/system-wright Skill
0★ repoTurns a vague goal, messy repeated workflow, AI adoption problem, or product/operations idea into a reusable AI work system built from Prompt, Context, Harness, and Loop, with a plain-language brief, human-approval gates, and a small trial run. Use when the user wants to productize a recurring way of working with an agent (Codex, Claude Code, Gemini CLI, Cursor, or compatible) rather than get a single answer. Do not use for one-off rewrites, pure definitions, temporary checklists, or ordinary code review unless productizing a reusable workflow.
Abed-Zahed/claude-skills-kit Skill
no license0★A complete Claude Code skill kit — prompt polishing with auto-routing, 5-phase engineering workflow, web research, knowledge ingestion, advisory board simulation, and multi-agent focus group. Skills work standalone or as an integrated system.
Drvivek34/Skill-Bazaar/ai-ml/claude-api Skill
no license0★ repoReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).
The same file is in 1 other repository
drewid74/unified-token-telemetry Skill
no license0★Use this when: track AI token usage across providers, build unified cost dashboard, how much am I spending on AI, compare model costs, cache hit ratio monitoring, LiteLLM token tracking, GitHub Copilot usage, OpenAI usage API, Anthropic usage API, ChatGPT Plus cost tracking, Claude Pro cost tracking, Gemini Advanced cost, Perplexity Pro cost, Cursor usage, Windsurf usage, multi-provider token telemetry, subscription cost tracking, idempotent token collection, hourly token rollups
digitaldreams/tuhin/skills/api-contract Skill
0★ repoDesigns the REST API contract from tasks/requirements.md, tasks/system_design.md and tasks/system_model.md, writing tasks/api_contract.md. Use whenever the user says "api contract", "design the api", "api spec", "endpoint design", or as the SDLC pipeline step between system-modeling and task-breakdown.
goingli0324/ui-ux-deploy-reviewer Skill
0★部署前的 UI/UX 總體檢。以全球公認的 20 項 UI/UX 原則(Nielsen 十大易用性啟發法 + 互動心理學定律 + 可及性與適應性)逐項審視網站或 Web App 的前端程式碼與使用者體驗,輸出「依嚴重度排序的問題清單 + 修正建議 + 部署放行檢核表」。MANDATORY TRIGGERS:使用者說「部署前 UI/UX 檢查」「deploy 前跑一次 UI/UX」「上線前幫我檢查 UI/UX」「審查 UI/UX」「用 20 原則檢視」「這個網站好不好用」「易用性審查」「heuristic evaluation」「檢查使用者體驗」,或在準備 firebase deploy / 上線 / 發布前要求做最後檢查時,都要套用此 skill。即使使用者沒明說「UI/UX」,只要意圖是部署前對介面與體驗做整體審視,就觸發。SCOPE:本 skill 審查 UI/UX 與呈現層程式碼品質;資訊安全審查請改用 web-security-reviewer。本 skill 同時是 agentic-dev-loop「Verify 雙閘」的 UI/UX 閘。分流原則:泛泛的「部署前/上線前檢查」(未指明只要 UI/UX)應交給 agentic-dev-loop 走雙閘,以免漏掉安全審查;本 skill 只在意圖明確聚焦 UI/UX/易用性/20 原則時觸發。
fpizzuta/repo-bootstrap Skill
0★Clone a GitHub repo, read its own documentation, find and fill the gaps in the local environment, install the project, and verify it actually runs. Use this skill whenever the user gives a GitHub or git URL and wants it set up, installed, running, or tried out locally, and also when they ask "what do I need to run this", "get this working on my machine", "can you set this up", "clone and configure this", or paste a repo link with any intent to use it rather than just read it. Covers Python, Node/TypeScript, and Docker projects on Windows/PowerShell and on Linux containers such as Cowork.
qalamio/qalamlist-skills/skills/qalamlist-ops-companion Skill
0★ repoUse when working in the QalamList repo and the user asks to deploy, install prerequisites, check setup, run locally, configure Cloudflare, D1, Turnstile, secrets, Wrangler, diagnose deploy errors, or get QalamList running. Use for phrases like "deploy this", "doctor", "prerequisites", "publish", "setup", or "run locally".
shreytam/okf-toolkit/skills/okf-author Skill
0★ repoAuthor conformant Open Knowledge Format (OKF) v0.1 bundles — Google's open markdown+YAML-frontmatter spec for packaging organizational knowledge (tables, datasets, metrics, APIs, runbooks, playbooks) so AI agents and humans can consume it. Use this skill whenever the user mentions OKF, Open Knowledge Format, knowledge bundles, an "LLM wiki", agent-readable documentation, or wants to convert schemas, data dictionaries, catalogs, runbooks, internal docs, or a codebase (services, modules, API endpoints, architecture decisions) into a portable markdown knowledge base — even if they don't say "OKF" explicitly.
kartikmehta8/linear-to-pr/skills/ship-loop Skill
no license0★ repoDaily ticket-to-PR workflow. Use for ANY task involving Linear tickets, implementing features, commits, branches, or pull requests. Also shapes ALL output for an ADHD reader: lead with the next action, number steps, restate state every turn, no preamble or closing pleasantries. Trigger on every /ship run and on any coding, debugging, or planning message.
BASILAHAMED/claude-fable-skills Skill
no license0★Prompting and scaffolding guidance for Claude Fable 5 workflows. Use when Codex needs to create or revise prompts, skills, agent instructions, or harness behavior for Claude Fable 5 or Claude Mythos 5, especially for long-running agents, high-effort tasks, strong instruction-following, memory systems, subagent orchestration, repo work, and user-facing progress patterns.
the-c0d3r/claude-skill-memory-promotion/memory-promotion Skill
0★ repoReconciles Claude Code's CLAUDE.md, auto-memory (MEMORY.md and topic files, both personal and per-project), and any web-chat memory notes into one consistent, non-duplicated set of instructions. Use when the user asks to "clean up memory", "review my memory", "sync memory", "promote this to CLAUDE.md", "what's in my memory", or wants a periodic maintenance pass across memory stores.
johnisanerd/claude-skill-google-jobs-scraper/apify-google-jobs-scraper Skill
no license0★ repoRun a google jobs scraper that bills per listing, not per page, with the Apify Google Jobs Scraper API Pay Per Result Actor (johnvc/google-jobs-scraper---pay-per-result). Give a job title plus an optional location and get one flat row per opening with title, company_name, location, via (source platform), full description, job_highlights, detected_extensions (posted_at, schedule_type), apply_options direct links, source_link, and job_id. Built-in company_filter and via_filter drop unwanted employers or job boards before you pay for them. Use when the user asks for a google jobs scraper, wants to scrape Google Jobs openings for a role or city, needs a job listings dataset with predictable cost, or wants to exclude staffing agencies from the results. Pay-per-result billing, MCP-ready for Claude and other AI agents.
jpoindexter/design-case-study-skills/skills/ai-agent-case-study Skill
0★ repoStructure and write design case studies for AI agents, copilots, autonomous workflows, multi-agent systems, and other probabilistic products. Use when a case study must explain autonomy levels, permissions, approval thresholds, steering, observability, user-visible rationale, uncertainty, trust calibration, streaming, latency, orchestration, failures, retries, rollback, state recovery, evals, or human-in-the-loop design.
bradmccloskey/claude-cisco-skills/cisco-backup Skill
0★ repoGenerate Python scripts for Cisco device backup, inventory collection, and configuration management. Use when the user wants to back up configs, collect inventory, diff configurations, or track config changes.
alyshahendo/dbt-debugger/.claude/skills/dbt-debugger Skill
0★ repoDiagnose a failed dbt run/build/test by finding the model that ACTUALLY broke versus the ones that were only skipped as collateral. Use when a `dbt run`, `dbt build`, or `dbt test` fails; when the user shares dbt artifacts (manifest.json / run_results.json / a target/ directory); or when they ask "why did my dbt build fail", "what actually broke", "which model is the root cause", or "what's the blast radius". Produces an interactive lineage HTML with your root-cause analysis embedded in it.
itzTiru/claude-code-skills-fable5/cost-lean-orchestration Skill
no license0★ repoToken-efficient model routing for complex engineering sessions. The main model (Opus / Fable, high reasoning) never bulk-reads or explores — it delegates all exploration, bulk file reading, codebase search, log scanning, dependency tracing, and documentation reading to Sonnet subagents that return compact structured findings, and it spends its own tokens only on synthesis, architecture, judgment, and final code. Use this skill in EVERY session that involves reading more than a couple of files, exploring or mapping a codebase, investigating a bug, auditing call sites, reviewing a large diff, digesting logs or docs, or any multi-step engineering task — even if the user never mentions cost, tokens, subagents, or delegation. This is a standing operating discipline, not an on-request feature.
Apify google ai overview monitoring
johnisanerd/claude-skill-google-ai-overview-monitoring/apify-google-ai-overview-monitoring Skill
no license0★ repoMonitor whether a brand or domain is cited in Google AI Overviews with the Apify Google AI Overview API Actor (johnvc/Google-AI-Overview-API). Run a watchlist of queries on a schedule, check ai_overview_present, match your domain against the cited references, and build a citation history keyed by the fetched_at timestamp for AEO and GEO reporting. Use when the user wants google ai overview monitoring or tracking, brand visibility in AI Overviews, AI search visibility checks for a brand or client, answer engine optimization (AEO) or generative engine optimization (GEO) measurement, competitor citation share in AI Overviews, or alerts when a domain drops out of an AI Overview. Pay-per-retrieval billing, MCP-ready for Claude and other AI agents.
Apify app store reviews scraper
johnisanerd/claude-skill-app-store-reviews-scraper/apify-app-store-reviews-scraper Skill
no license0★ repoScrape Apple App Store reviews into structured JSON with the Apify Apple App Store Reviews API Actor (johnvc/apple-app-store-reviews-api). Give it a numeric Apple product ID or just an app name and get one row per review with rating, review_title, review_text, author_name, reviewed_version, review_date_iso, and app context, across 52 country stores, sorted by most recent, most helpful, most favorable, or most critical. Use when the user wants an app store reviews scraper, wants to scrape App Store reviews or iOS app reviews, export Apple app reviews to JSON or CSV, build a review dataset for one or more apps, or pull the reviews behind a competitor's App Store listing. Pay-per-review billing, MCP-ready for Claude and other AI agents.
marian-kamenistak/engineering-salary-calculator Skill
0★Use when estimating a software developer's or engineering manager's market value or salary, benchmarking pay against 2026 European market data, preparing a raise or negotiation case, or scoring engineering and leadership skills to find what's holding a level up.
mightykevster/mightyharness Skill
0★Harness design — take a goal or function and produce a reviewable DESIGN DOCUMENT for the scaffolding around whatever LLM-driven process executes it: context, tools, guardrails, retry/verification loop, and executor topology. The artifact designs; it does not execute anything and never becomes the running system. Trigger on /harness, "design a harness", "scaffold this for an LLM to run", or any goal-directed LLM task to be engineered with harness-level rigor — not only software, but analysis, retrieval, or reasoning tasks. It has one job: output the best possible harness design.
rjvim/ai-skills/skills/rjv-codex-ollama-subagents Skill
0★ repoUse for any local-Ollama delegation: (a) Codex native subagents with repo tools — GPT-5.5 orchestrator, qwen/gemma explorers for read-only exploration, qwen/gemma workers for scoped edits, hybrid-ollama profile, Ollama serve tuning, mixed OpenAI-mini + local routing; and (b) one-shot local text generation (no repo tools) via the bundled ollama-chat.sh runner — draft-from-spec, tests, classify, summarize. Triggers: 'Codex local Ollama subagents', 'hybrid-ollama', 'qwen-explorer', 'gemma-explorer', 'qwen-worker', 'gemma-worker', 'ollama-chat', 'one-shot local model', 'draft from spec locally', 'save Codex tokens with local models'.
bispeklolik/debug-by-using/skills/debug-by-using Skill
0★ repoDebug by Using — универсальная методология отладки и закалки ЛЮБОГО приложения/программы через прямое ПРОЖИВАНИЕ (реальный запуск, весь пользовательский цикл, все фичи, как самый придирчивый нестандартный юзер), а НЕ через закрытые/зелёные тесты. Ловит баги, которых юнит-тесты со стабами не видят: роутинг, UX, экономика/учёт, состояние/связность, интеграции, прогрессия. Начинается с СТАРТОВОГО ИНТЕРВЬЮ, чтобы определить охват (всё или отдельные фичи), зафиксировать ЦЕЛЬ сессии (условие «готово», что гонит работу до конца) и ЭТАЛОН (идеал, по которому судим «хорошо»). ДВА РЕЖИМА: «Соло» (сам проживаешь и чинишь инлайн) и «Агентик+» (делегируешь фиксы агентам ПАРАЛЛЕЛЬНО). Применимо к веб- и десктоп-UI, CLI, API, ботам, играм, пайплайнам, прошивкам — к любому продукту с пользовательским циклом. Включай, когда пользователь говорит «debug by using», «проживи приложение», «отладь использованием», «пройди весь цикл», «используй все фичи», «проверь на прочность живьём», «поработай как реальный юзер», «в соло-режиме»/«в агентик-режиме», «не тестами, а использованием»; ИЛИ тесты зелёные, но реальное поведение не проверено; ИЛИ перед релизом/демо/продажей; ИЛИ «всё работает», но чувствуешь, что живьём отвалится.
plori-ai/skills/plori Skill
0★ repoCreate and drive plori agents (each an AI agent on its own cloud computer) from any MCP client, the plori CLI, or over REST. Covers authentication (OAuth 2.1 or API key), creating agents, invoking them and reading replies, answering human-in-the-loop requests, and scheduling deferred runs.
kourika/Biz-PPT-Doctor Skill
0★Content diagnosis and audit coach for workplace reports and business proposal presentations (职场汇报, 商业提案, 述职, 项目复盘). Use this skill whenever the user wants to turn raw material (meeting notes, data, reports) into a logically structured outline BEFORE building slides; wants an existing .pptx/.ppt reviewed, audited, diagnosed, or '体检' for logic, design consistency, data credibility, or Chinese typography problems; needs to triage a pile of reviewer or boss feedback (老板意见/评审意见/导师意见) into concrete per-slide fixes; needs help choosing the right chart type for data; or needs a speaker script matched to a deck's timing. Trigger on phrases like '这个PPT怎么样' '帮我看看这份PPT' 'PPT逻辑不清楚' '大纲怎么搭' '老板给了一堆意见改不过来' '这页用什么图表' — even if the user never says 'audit' or 'diagnose' explicitly. This skill owns CONTENT and STRUCTURE, not file generation: pair it with Anthropic's built-in pptx skill (or another pptx-generation skill) to actually create or edit the .pptx file.
ahmadabdulnasir/agent-ai-readiness/skill/agent-web-audit Skill
0★ repoAudit a website or web app for AI-agent readiness, accessibility, and AI-search legibility, then fix the issues found. Use this skill whenever the user wants their site to work well for AI agents or LLMs (Claude, ChatGPT, Perplexity, Grok, Gemini, browser agents), or mentions "agentic SEO", "GEO", "AEO", "AI-friendly site", "agent-ready", "accessibility tree", "llms.txt", "make my site readable for AI", or the Lighthouse "Agentic Browsing" audit. Also trigger when the user asks to check or fix a frontend for semantic HTML, ARIA, label linking, cursor:pointer, layout stability, or how a site appears to crawlers and agents. Works on plain HTML, React/Next.js, Vue, and server-rendered templates. Trigger even if the user only says "make this site useful for AI" without naming a specific check.
fibrousop07-dev/lazy-skills Skill
0★Routes multi-step or multi-tool tasks to the right combination of connected skills without loading every skill's full instructions upfront. Use this whenever a task plausibly touches more than one specialized skill in sequence or in combination — e.g. "query the database and validate the schema," "format this SQL and check it against the schema," or any request where completing it well would mean chaining two or more skilled capabilities together. Also use it proactively whenever you're about to invoke a skill and aren't sure whether related skills exist that the task also needs. Do NOT use this for single-skill tasks that already trigger cleanly on their own — it adds a lookup step that isn't worth it there.
zenlee123/routerbase-agent-skills/skills/routerbase-api-integration Skill
no license0★ repoIntegrate applications with RouterBase, the OpenAI-compatible model gateway at https://routerbase.com/v1. Use when migrating OpenAI SDK calls to RouterBase, configuring RouterBase API keys, implementing chat completions, streaming, tool calling, JSON mode, vision inputs, request validation, error handling, or documenting RouterBase setup for Python, JavaScript, curl, LangChain, LlamaIndex, Vercel AI SDK, Cursor, Continue, or other OpenAI-compatible clients.
The same file is in 2 other repositories
Apify google maps directions api
johnisanerd/claude-skill-google-maps-directions-api/apify-google-maps-directions-api Skill
no license0★ repoGet Google Maps directions as structured JSON with the Apify Google Maps Directions API Actor (johnvc/google-maps-directions-api). Give a start and end point as an address, GPS coordinates, or a Google Maps place ID, pick a travel mode (best, driving, cycling, walking, transit, flight, two-wheeler), and get route options back with formatted_distance, formatted_duration, a via summary, turn-by-turn steps, a per-mode durations summary, resolved place coordinates, and a google_maps_directions_url. Use when the user wants a google maps directions api, route data as JSON, distance and ETA between two addresses, turn-by-turn directions inside a script or agent, or directions without a Google Cloud API key. Pay-per-route billing, MCP-ready for Claude and other AI agents.
johnisanerd/claude-skill-duckduckgo-scraper/apify-duckduckgo-scraper Skill
no license0★ repoScrape DuckDuckGo search results into structured JSON with the Apify DuckDuckGo Scraper Actor (johnvc/DuckDuckGo-Scraper-for-serp-rankings). One query returns a page item holding organic_results rows with position, title, link, snippet, favicon, and a date when DuckDuckGo shows one, plus page metadata and slots for ads, knowledge graph, news, inline images, inline videos, and related searches. Localize across 42 region and language codes, filter by date, set safe search, and cap pages to control spend. Use when the user wants a duckduckgo scraper, duckduckgo scraping, DuckDuckGo search results as JSON or CSV, a DuckDuckGo SERP export, or a keyless way to read DuckDuckGo from code. Pay-per-page billing, MCP-ready for Claude and other AI agents.
Apify app store optimization reviews
johnisanerd/claude-skill-app-store-optimization-reviews/apify-app-store-optimization-reviews Skill
no license0★ repoMine Apple App Store reviews for app store optimization signals with the Apify Apple App Store Reviews API Actor (johnvc/apple-app-store-reviews-api). Pull most-critical and most-recent reviews for your app and competitor apps across 52 country stores, then turn the rows into complaint themes, rating trends by reviewed_version, and per-country feedback that feeds listing updates and release notes. Use when the user wants app store optimization insight from reviews, an ASO review audit, competitor app review analysis, review sentiment for an iOS app, to find out why users leave 1-star reviews, or to check how an app update changed the reviews. Review mining only; keyword rankings belong to the Apple App Store Search Actor and listing metadata to the Product API Actor. Pay-per-review billing, MCP-ready for Claude and other AI agents.
ayman-benmada/owasp-security-audit/skills/security-audit Skill
0★ repoPerforms a complete application security analysis based on the OWASP Top 10 (2025). Use this skill whenever the user mentions: security audit, OWASP analysis, security code review, application vulnerabilities, application pentest, security review, security flaw, API security, code verification, hardening, or asks to verify the security of code, an architecture, or a configuration. Also trigger for terms such as: SQL injection, XSS, CSRF, weak authentication, secrets in code, sensitive data exposure, access control, SSRF. This skill orchestrates 10 OWASP category reference guides and produces a structured, actionable report.
SeamStressDev/seam-scaffold/skills/fixture-immutability Skill
0★ repoUse this protocol whenever a test fails after a change, whenever editing test files, test fixtures, expected outputs, snapshots, or scoring criteria, and whenever the temptation exists to make a red check green by touching the check instead of the code.
babble-open-source/helpnest-skills/skills/helpnest-author Skill
no license0★ repoAuthor or update customer-facing help-center articles in HelpNest via its MCP server. Use whenever the user wants to write, publish, refresh, or fix help articles, help-center docs, knowledge-base content, support articles, or FAQ content for HelpNest — or when they ask "what should I document", "close the learning loop", or "what are customers asking that we can't answer" (knowledge gaps). Handles workspace verification, collection lookup/creation, idempotent create-or-update, draft-vs-publish decisions, code-grounded UI labels, and concurrency-safe editing. Trigger this even when the user only names the article topic (billing, notifications, onboarding) without saying "HelpNest" or "help article", as long as the destination is the HelpNest help center. Not for developer/engineering docs, READMEs, API references, or internal wikis.
johnisanerd/claude-skill-hotel-price-monitoring/apify-hotel-price-monitoring Skill
no license0★ repoRun hotel price monitoring on Google Hotels with the Apify Google Hotels Search Scraper Actor (johnvc/google-hotels-search-scraper). Poll the same destination and stay dates on a schedule, snapshot rate_per_night and total_rate per property with numeric extracted values, then diff snapshots keyed on property_token to alert on drops, watch competitor hotel rates, or chart how prices move as the stay date approaches. Use when the user wants hotel price monitoring, wants to track or watch hotel prices over time, monitor competitor hotel rates for a property or market, build a rate drop alert pipeline, or asks whether hotel prices go down closer to the date. Pay-per-page billing plus a small per-run setup fee, MCP-ready for Claude and other AI agents.