Test
10,858 rows from 2,054 repositories
Proving it works, and keeping it that way.
What Test skills agree on
1,559 skill files read, by 1,201 of the 2,096 authors on this shelf whose files we hold, 2026-09-06
The middle one of the 175 measured here is ~1.7k tokens long, counted with cl100k_base
Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.
What they tell the agent to do
- Write a failing test before writing code43 of 1201 in 36 filesalso worded as write failing tests before implementing code; Write a failing test first
- Run the full test suite36 of 1201 in 35 filesalso worded as run the full test suite fresh; run full test suite
- Test only one variable per experiment34 of 1201 in 17 filesalso worded as test only one variable per test; Change only one variable per test
- Read product marketing context before asking questions34 of 1201 in 14 filesalso worded as Check product marketing context before asking questions; read project context before asking questions
- Mock external dependencies34 of 1201 in 30 filesalso worded as Mock all external dependencies; Mock external dependencies appropriately
- Define primary, secondary, and guardrail metrics33 of 1201 in 16 filesalso worded as define primary secondary and guardrail metrics; Define primary and guardrail metrics before testing
- Pre-determine sample size before starting31 of 1201 in 14 filesalso worded as pre-determine sample size before testing; Pre-determine sample size before launching
- Test behavior rather than implementation31 of 1201 in 29 filesalso worded as Test behavior rather than implementation details; Test behavior and outcomes instead of implementation details
- Formulate a hypothesis before designing a test30 of 1201 in 13 filesalso worded as Formulate a specific hypothesis before testing; Formulate a single hypothesis before testing
- Document every test hypothesis, variant, and result29 of 1201 in 11 filesalso worded as Document hypothesis, variants, results, and learnings; document every test with hypothesis variants and results
- Use descriptive test function names25 of 1201 in 21 filesalso worded as Use descriptive names for all test functions; use descriptive names for all tests
- Commit to the methodology without stopping early24 of 1201 in 8 files
- Verify tracking and complete QA before launch24 of 1201 in 9 filesalso worded as verify tracking and implementation before launch
- Test happy paths and edge cases22 of 1201 in 20 filesalso worded as Test both happy paths and error cases; include happy path and edge case scenarios
- Test edge cases and boundary conditions21 of 1201 in 15 filesalso worded as cover edge cases and boundary values; Test edge cases and error conditions
What they tell it not to do
- Do not peek at results and stop early44 of 1201 in 26 filesalso worded as do not stop tests early due to good results; Do not peek at results before the test concludes
- Do not test implementation details43 of 1201 in 33 filesalso worded as do not test implementation internals; Do not test internal implementation details
- Do not skip tests35 of 1201also worded as do not skip manual smoke tests; never skip tests without an active issue reference
- Do not make changes to variants during the test34 of 1201 in 15 filesalso worded as do not make changes to variants mid-test; Do not make changes to variants during a test
- Do not add traffic from new sources mid-test30 of 1201 in 13 filesalso worded as do not add traffic from new sources during the test; Do not add traffic from new sources during tests
- Do not share state between tests29 of 1201 in 23 filesalso worded as Do not share mutable state between tests; Do not share mutable state across tests
- Do not test private methods27 of 1201also worded as Do not test private methods directly; do not test private functions directly
- Do not ignore test failures26 of 1201 in 20 filesalso worded as Do not ignore flaky tests; Do not ignore errors
- Do not cherry-pick segments during analysis20 of 1201 in 6 filesalso worded as Do not cherry-pick segments; do not cherry-pick segments after the fact
- Do not modify source files19 of 1201 in 16 filesalso worded as Do not modify any files; do not modify source code
What they expect to be installed
- git215 of 1201
- pytest130 of 1201 in 127 files
- grep102 of 1201
- curl78 of 1201 in 72 files
- Playwright73 of 1201 in 72 files
- vitest70 of 1201 in 65 files
- npm66 of 1201 in 62 files
- bash60 of 1201
- gh60 of 1201 in 56 files
- python350 of 1201
What they ask it to produce
- Coverage report70 of 1201also worded as coverage reports; code coverage reports
- Test results68 of 1201 in 57 filesalso worded as Test execution results; test results analysis
- Screenshots60 of 1201 in 55 filesalso worded as browser screenshots; screenshot comparisons
- Test files54 of 1201 in 51 filesalso worded as Test file; pytest test files
- Unit tests51 of 1201 in 46 filesalso worded as Unit test files; Unit test classes
- Integration tests42 of 1201 in 37 filesalso worded as Integration test files; Integration test results
- Test report42 of 1201also worded as Test execution report; Test summary report
- Test plan41 of 1201 in 36 filesalso worded as Test plan document; test plans
When Test authors say to reach for one
The situations these authors wrote into their own files, counted out of the same 1,201 authors, with the skills that name each one
- User asks to write tests52 of 1201
- Test quality
- Assay
- Publish to cloud
- Creative testing
- and 1 more on this shelf
- Writing tests47 of 1201 in 42 files
- Test master
- Senior qa
- Testing
- Flutter dev
- Qa engineer
- Senior qa skills davila7 senior qa
- and 13 more on this shelf
- User requests to run tests40 of 1201 in 37 files
- A/B test34 of 1201 in 19 files
- Prompt governance
- Ads test
- Ab testing
- A b testing
- Growth experimentation
- Ab test setup
- and 19 more on this shelf
- Test this change33 of 1201 in 17 files
- Ab testing
- Ab test setup
- Marketing ab testing
- and 14 more on this shelf
- User asks to check test coverage29 of 1201
- Variant copy28 of 1201 in 12 files
- Ab testing
- Ab test setup
- Marketing ab testing
- and 14 more on this shelf
- Experiment27 of 1201 in 15 files
- Ads test
- Ab testing
- A b testing
- Growth experimentation
- Ab test setup
- Marketing ab testing
- and 13 more on this shelf
How Test skills are built
9,751 skill directories by 1,851 authors, read from their repositories’ own file trees 2026-08-05
The middle bundle among those shipping files is 3 files, 16.9 KB beside SKILL.md
Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.
The shape
- SKILL.md is the whole skill668 of 1,851 authors, 4,943 of 9,751 skills
- files ship beside it1,183 of 1,851 authors, 4,808 of 9,751 skills
- executable scripts ship inside462 of 1,851 authors, 1,392 of 9,751 skills
The folders they converge on
- references/723 of 1,851 authors, 2,373 of 9,751 skills
- scripts/367 of 1,851 authors, 1,170 of 9,751 skills
- agents/215 of 1,851 authors, 584 of 9,751 skills
- assets/154 of 1,851 authors, 423 of 9,751 skills
- examples/132 of 1,851 authors, 276 of 9,751 skills
- evals/109 of 1,851 authors, 275 of 9,751 skills
microsoft/playwright-mcp/io.github.microsoft/playwright-mcp MCP server
in 1 stack36,126★ repoPlaywright Tools for MCP
shanraisshan/claude-code-best-practice/.claude/skills/agent-browser Skill
64,157★ repofrom vibe coding to agentic engineering - practice makes claude perfect
wshobson/agents/plugins/accessibility-compliance/skills/screen-reader-testing Skill
38,701★ repoMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLI
keploy/keploy/io.github.keploy/mcp MCP server
18,384★ repoEnd-to-end API testing — generate and run tests from OpenAPI, curl, Postman, or real user traffic.
citrolabs/ego-lite/skills/ego-browser Skill
9,516★ repoThe fastest browser for AI agents to run web automation, built for sharing your logged-in browser state with your AI agents, like Codex or Claude Code, without disturbing you. Zero cost, zero config.
lackeyjb/playwright-skill/skills/playwright-skill Skill
3,026★ repoClaude Code Skill for browser automation with Playwright. Model-invoked - Claude autonomously writes and executes custom automation for testing and validation.
noobnooc/agent/skills/better-codex Skill
no license1,374★ repoMy profile & the agent skills I created
AI-Builder-Club/skills/skills/crabbox-setup Skill
no license1,165★ repoCodebase harness + loop engineer
countbot-ai/CountBot/workspace/skills/agent-browser Skill
762★ repo更适配中文用户的轻量开源AI Agent | 国产大模型Coding plan支持 | 兼容OpenClaw Skills生态| 已接入微信ClawBot/微博龙虾/飞书/钉钉/QQ/小智AI/Telegram/deepseek-v4。
indranilbanerjee/digital-marketing-pro/skills/ab-test-plan Skill
734★ repoOpen-source AI marketing plugin for agencies & in-house teams — 158 skills, 25 specialist agents, 12-Part Strategy Flow, Cowork team-persistent, EU AI Act Article 50 ready, 6-platform AEO/GEO incl. Google AI Mode. Installs on Claude Code, Cowork, Codex, Cursor, Copilot CLI, Antigravity. MIT-licensed.
levnikolaevich/claude-code-skills/plugins/codebase-audit-suite/skills/ln-21-documentation-auditor Skill
535★ repo18 standalone skills for Claude Code and Codex: review, audit, optimization, testing, product discovery, and safe repository publishing.
VisionForge-OU/foreman/src/foreman/skills/foreman-debug Skill
no license484★ repoA Boris-style agentic orchestrator TUI that supervises headless Claude Code agents through a gated software-delivery pipeline — pointed at any repository.
mxyhi/ok-skills/agent-browser Skill
472★ repoCurated AI coding agent skills and AGENTS.md playbooks for Codex, Claude Code, Cursor, OpenClaw, and other SKILL.md-compatible tools.
SponsioLabs/Sponsio/plugins/sponsio-claude-code/skills/configure Skill
445★ repoDeterministic safety solutions for probabilistic AI agents
AvdLee/Swift-Testing-Agent-Skill/swift-testing-expert Skill
435★ repoAn agent skill focused entirely on Swift Testing, helping you write better tests, migrate from XCTest, improve test architecture, and adopt modern Swift testing patterns with confidence.
Eyadkelleh/awesome-skills-security/skills/llm-testing Skill
no license353★ repoSecurity testing toolkit for AI Agent: curated SecLists wordlists, injection payloads, and expert agents for authorized pentesting, CTFs, and bug bounties
ai-dashboad/flutter-skill/io.github.ai-dashboad/flutter-skill MCP server
351★ repoAI-powered E2E testing for 10 platforms. 253 MCP tools. Zero test code needed.
testdino-hq/playwright-skill Skill
349★TestDino Playwright Skill: AI-powered guides for Playwright best practices, made by testdino.com.
Microck/ordinary-claude-skills/skills_all/activitypub-testing Skill
no license324★ repoAn unappealing collection of Claude Skills and resources.
microsoft/GitHub-Copilot-for-Azure/.github/skills/analyze-skill-issues Skill
no license242★ repoGitHub Copilot for Azure
Designing distributed system tests
shenli/distributed-system-testing/skills/designing-distributed-system-tests Skill
226★ repoAI-agent skills for distributed-systems testing
aigorahub/elves Skill
203★Autonomous multi-batch development skill for Claude Code and Codex. They work while you sleep.
workersio/skills/plugins/wio/skills/wio Skill
155★ repoAgent skills to find and fix software bugs
serpro69/claude-toolbox/cmd/generate-kodex/testdata/source-plugin/skills/test-skill Skill
no license148★ repoMinimal by design, explicitly multi-lang, production-ready and battle-tested collection of configs and plugins for your Claude Code agentic development workflows: mcp, configs, skills, agents and more.
RayFernando1337/rayfernando-skills/plugins/bootstrap-ios/skills/bootstrap-ios Skill
119★ repoRay Fernando's collection of Skill files for AI coding agents. First up: running-bug-review-board, a real-user QA workflow with a Bug Review Board (BRB) feedback loop.
avibebuilder/claude-prime/.claude/skills/agent-browser Skill
116★ repoYou've heard Claude Code can do amazing things. Skills, hooks, agents, memory systems — but who has time to figure all that out? Claude Prime sets it up for you in one command.
incogbyte/android-reverse-engineering-claude-skill/plugins/android-reverse-engineering/skills/android-reverse-engineering Skill
98★ repoClaude Code skill that automates Android application reverse engineering
bocato/swift-testing-agent-skill/swift-testing Skill
78★ repoAgent Skill providing expert Swift Testing guidance for AI coding tools: covering test doubles, fixtures, async patterns, XCTest migration, and testing best practices.
geekjourneyx/travel-guidebook Skill
75★AI agent skill that generates beautifully typeset travel guidebook PDFs — from research to print, powered by parallel agents and Amap MCP.
ominou5/funnel-architect-plugin/skills/ab-testing Skill
73★ repoTurn Claude into a full-stack sales funnel architect. Build high-converting funnels with optimized UI/UX, page speed, mobile responsiveness, and deploy to any platform.
petrkindlmann/qa-skills/skills/accessibility-testing Skill
70★ repo50 QA and test-automation skills for Claude Code, Codex, Cursor, and any Agent Skills Standard runtime.
adriannoes/awesome-agentic-ai/cursor-claude-codex/skills/alirezarezvani-skills/playwright-pro/skills/browserstack Skill
48★ repo329 agent skills (Cursor, Claude Code & Codex), 5,380 OpenClaw skills, 201 ML notebooks, 7 textbooks, 52 research papers, 17 industry reports for PMs, Designers & Developers.
Stanestane/game-design-skills-bundle/design-red-team-audit Skill
no license44★ repoA public OpenClaw and ClawHub-ready bundle of game design skills for emotional direction, feature workflows, audits, FTUE, pitch decks, prototyping, and red-team design review.
fcavalcantirj/claude-code-eyes Skill
43★Give Claude Code eyes 👁️ — a camera skill so it can SEE real hardware, displays and wiring: verify a rendered panel, check wiring before power-on, and catch bugs that live on the glass, not the logs.
goldbergyoni/test-coverage-mcp/io.github.goldbergyoni/test-coverage-mcp MCP server
no license41★ repoAnalyze test coverage from LCOV files - makes AI agents coverage-aware without wasting tokens
cypress-io/ai-toolkit/skills/cypress-author Skill
38★ repoFast, flexible, and open tooling for building intelligent workflows with Cypress.
qase-tms/qase-mcp-server/io.qase/mcp-server MCP server
31★ repoOfficial MCP server for Qase — manage test cases, runs, suites, defects via AI tools.
Shelpuk-AI-Technology-Consulting/agent-skill-tdd/skills/tdd Skill
30★ repoTDD + requirements workflow skill for AI coding agents (Claude Code, Codex, Antigravity, Cursor, Windsurf, etc.). Part of the Shelpuk agentic suite alongside Serena, Kindly Web Search, and Lad.
inbharatai/claude-skills/skills/ab-test-designer Skill
28★ repoCurated Claude and Codex skill collection for local AI workflows, automation, research, and product-building.
Dotnet testing advanced aspire testing
kevintsengtw/dotnet-testing-agent-skills/skills/dotnet-testing-advanced-aspire-testing Skill
28★ repoAI Agent Skills for .NET Testing - Based on 30-Day Testing Challenge (iThome Ironman 2025 Winner)
mralaminahamed/wp-dev-skills/skills/wp-admin-browser Skill
27★ repoWordPress plugin development skills for AI coding agents — Claude Code, Gemini CLI, Cursor, Windsurf, Cline, Codex, Copilot, opencode, and more
mjunaidca/mjs-agent-skills/.claude/skills/browsing-with-playwright Skill
no license27★ repoA curated collection of Agent Skills — reusable units of intelligence that teach AI General Agents how to perform specific tasks autonomously.
jbrazy480/loop-maker/loop-maker Skill
25★ repo🔁 Write loops, not prompts. Free skill that generates overnight autonomous AI runs for Claude Code, Codex & Cursor: master prompt, plan, verifier, one command. By AI Guy Official.
oribarilan/97/.todo/done/US-97-mvp/reference Skill
no license23★ repoAgent skills distilled from the hard-won lessons of world-renowned programmers, in the spirit of "97 Things Every Programmer Should Know"
Orizon-eu/claude-code-pentest/api-breaker Skill
22★ repo6 Claude Code skills that automate the entire pentest lifecycle. From recon to exploit chains to bug bounty reports — just give it a domain. 43 scripts, zero dependencies, pure Python.
igmarin/rails-agent-skills Skill
22★This is my personal configuration of skills as a Ruby on Rails Dev
knot0-com/vibe-testing Skill
22★Pressure-test your specs with LLM reasoning before writing code. Agent skill for Claude Code, Codex, Gemini CLI, and 14+ coding agents.
marceloeatworld/nixos-ai-skill Skill
no license20★Auto-updated NixOS & Nix ecosystem documentation for AI coding assistants — works with 33+ tools via the Agent Skills standard (SKILL.md)
galyarderlabs/galyarder-framework/integrations/antigravity/ab-test-setup Skill
20★ repoThe framework that powers your Agentic Company
martinemde/skillet/.claude/skills/task-test Skill
19★ repo🍳 Run SKILLs as Shell Scripts. Cook with Claude without the mess.
TheStack-ai/pulser/skill Skill
18★ repoSKILL.md linter for Claude Code — auto-diagnose and fix skills against Anthropic best practices. npx pulser
chrisallenlane/claude-swe-workflows/skills/bug-fix Skill
18★ repoSoftware Engineering workflows for Claude Code (MIRROR)
Chipagosfinest/enterprise-team/enterprise-team/skills/data-catchall Skill
no license18★ repoHire a whole company with one plugin. 75 specialized AI agents across Engineering, Product, Infrastructure, Data, Security, Marketing, Sales, Finance, Legal, and People.
livlign/claude-skills/plugins/dotnet-coverage-kit/skills/coverage-init Skill
18★ repoA Claude Code plugin marketplace for open-source maintainers — README heroes, README audits, and .NET test-coverage backfills.
WenOwen/wu5-dev-flow/wu5-dev-flow Skill
18★ repoAuditable SDD, strict TDD, and safe Git workflow skill for Codex and Python projects.
TradeOS-AI/tradeos-mcp/io.github.TradeOS-AI/tradeos-mcp-test MCP server
18★ repo[TEST] TradeOS MCP bridge: ticker search, My Agent, chart TA, macro news. npm or HTTP.
OrangeproAI/orangepro-mcp/io.github.OrangeproAI/orangepro MCP server
17★ repoFind test gaps, generate grounded tests, and dynamically prove behavior with mutation testing.
alex-feel/claude-code-toolbox/tests/e2e/fixtures/mock_repo/skills Skill
16★ repoClaude Code Toolbox — automated installers and environment configuration framework for Claude Code with one-line setup across Windows, macOS, and Linux.
prog-time/claude-config-template/skills/hello-test Skill
no license16★ repoA template repository for your personal Claude Code configuration
joeynyc/skillscore/test-skill Skill
16★ repoCLI tool that evaluates AI agent skills and produces quality scores. Works with any SKILL.md-based skill from skills.sh, ClawHub, GitHub, or local directories.