mejbaurbahar
UserFagun — a QA/testing Skill that gives any AI agent a real browser to click, crawl, and find real bugs — functional, security, performance, accessibility, SEO. Every finding is backed by evidence. Nothing is faked; if it can't be reproduced, it isn't reported.
Categories
Indexed Skills (24)
accessibility-testing
Use when auditing a web app for accessibility (a11y) — WCAG compliance, screen reader compatibility, keyboard navigation, color contrast, and semantic HTML/ARIA correctness.
ai-ml-testing
Use when testing a classical ML model or ML-powered feature — accuracy/precision/recall/F1 evaluation, model regression testing, bias/fairness testing, and ML pipeline robustness. For LLM/generative-AI-specific testing (prompts, hallucination, RAG), use llm-testing instead.
api-testing
Use when testing REST/GraphQL APIs directly — status codes, schema/contract validation, auth, pagination, rate limits, idempotency, and error responses. Use instead of driving the UI when the question is really about the API layer.
autonomous-qa-orchestration
Use when asked to autonomously test an application end-to-end with minimal hand-holding — discover, explore, plan, execute, investigate, report, and verify fixes as one continuous QA pass. The top-level playbook that pulls in the other testing skills as needed.
bug-reporting
Use whenever filing a defect found during any kind of testing — applies the harness's standard severity scale, report template, and evidence requirements so reports are consistent and actionable.
chaos-resilience-testing
Use when deliberately injecting failure (network latency/loss, service/dependency crash, resource exhaustion) to verify a system degrades gracefully instead of cascading — retries, circuit breakers, timeouts, backpressure. Requires explicit authorization and a scoped blast radius; never run against unscoped production without sign-off.
chatbot-voice-agent-testing
Use when testing chatbots, voice assistants, or conversational AI agents specifically — multi-turn conversation flow, context/memory retention, intent recognition, STT/TTS accuracy, interrupt handling, and fallback behavior. Builds on llm-testing for the underlying model concerns.
cicd-testing
Use when designing or debugging test execution in CI/CD pipelines — PR gates, test tiering (smoke/regression/nightly), parallelization/sharding, quality gates, flaky-test tracking, and failure notification wiring. Covers GitHub Actions, GitLab CI, Jenkins and similar.
cloud-infrastructure-testing
Use when testing cloud/infrastructure concerns — Docker/Kubernetes deployments, serverless functions, managed-service integration (S3, queues, API Gateway), disaster recovery/failover, and infra-as-code correctness on AWS/Azure/GCP.
database-testing
Use when verifying data integrity, CRUD correctness at the persistence layer, migrations, transactions/concurrency, or query correctness/performance — SQL and NoSQL. Use whenever a bug claim needs to be confirmed against actual stored state, not just UI/API appearance.
fagun
Fagun — end-user simulation + User Acceptance Testing (UAT) + autonomous QA and real-bug hunting. Use whenever the user wants to know whether a website/web app is truly ready for real customers: use the product as multiple real personas (first-time, mobile, slow-internet, low-end, keyboard-only, screen-reader, international…), validate complete user journeys (register, login, search, checkout, password reset…) end to end, judge UX/UI/business-logic quality, AND find REAL reproducible defects (console/JS errors, failed requests, broken links, form-validation gaps, auth/session flaws, accessibility violations, performance regressions, security misconfig). Produces a product-readiness scorecard and a release verdict. Drives the `fagun` MCP browser tools. Triggers: "/fagun", "test this site", "UAT <url>", "is <url> ready for users", "review the UX of <url>", "find bugs on <url>", "deep test", "audit <url>", "readiness of <url>".
functional-testing
Use when verifying that a feature or user flow behaves according to requirements/specs — functional correctness testing of forms, workflows, business logic, and CRUD operations, whether done manually or via automation.
llm-testing
Use when testing LLM-powered features — prompt testing, hallucination/faithfulness checks, RAG evaluation, agent/tool-use testing, prompt injection and LLM security, regression testing across model/prompt versions, and cost/latency/token budget testing.
manual-testing
Use when designing test cases, doing exploratory testing, writing test plans/checklists, or manually verifying a feature before or instead of automating it. Covers test case design techniques, exploratory charters, and test plan structure.
mobile-testing
Use when testing mobile apps (iOS/Android native, or responsive/mobile-web) — device/viewport matrices, touch gesture testing, mobile-specific network and lifecycle conditions, and app-store readiness checks.
observability-testing
Use when correlating a test failure or production issue across logs, metrics, and traces to find root cause; when verifying that a feature emits the logging/metrics/alerting it needs; or when doing shift-right testing via production monitoring.
performance-testing
Use for load/stress/soak/spike testing, Core Web Vitals audits, Lighthouse checks, API latency benchmarking, and diagnosing slow pages or endpoints. Covers both frontend perf and backend load testing.
qa-code-intelligence
Use when reading source code to spot testability/risk issues before writing tests, generating test cases automatically from requirements/schemas/code, or doing root-cause/bug intelligence work — clustering failures, deduping reports, and reasoning from stack traces/logs to a likely cause.
qa-fundamentals
Use when scoping test strategy, writing a test plan, classifying testing types (QC vs QA, verification vs validation), applying risk-based prioritization, or navigating SDLC/STLC/Agile testing process questions. The conceptual root the other testing skills build on.
regression-testing
Use when validating that recent changes haven't broken existing functionality — building/running regression suites, deciding regression scope for a release, and impact analysis from a diff/changelog.
security-testing
Use for authorized security/penetration testing of web apps and APIs — OWASP Top 10 checks, auth/session testing, header/config audits, and vulnerability triage. Only for systems the user owns or has explicit authorization to test.
test-automation
Use when writing, structuring, or debugging automated tests — Playwright/Selenium E2E suites, pytest test design, page object models, CI test integration, and deciding what's worth automating.
test-data-engineering
Use when generating synthetic/realistic test data, seeding fixtures, masking PII for a test environment, or designing deterministic/reusable test datasets. Underpins reliable automation across all the other testing skills.
visual-testing
Use when checking visual/layout correctness — screenshot/visual regression testing, design-system and Figma-to-production validation, responsive layout, dark mode/theming, typography and spacing consistency. Distinct from functional correctness or accessibility (though related).
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.