Skip to main content
skill CX Codex

setup · fcakyon, qa — Claude Code skill: what it does, how to install

Skill for Claude Code configures real integration testing and user QA journeys across project codebases. It triggers when asked to set up QA or generate end-to-end testing flows.

Apache-2.0 ★ 1,216 updated 2026-11-16

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add fcakyon/claude-codex-settings
/plugin install qa@claude-settings
Open on GitHub

What setup does

This skill sets up end-to-end QA harnesses across websites, extensions, backends, Android projects, and macOS native apps. The workflow starts by running project-qa discover to inspect routes, package manifests, and existing test suites for candidate coverage.

Next, project-qa setup builds .agents/qa/config.json, .agents/qa/suite.mjs, and .agents/skills/qa/SKILL.md while leaving existing custom files intact. Generated placeholders are replaced with real user journeys and real fixtures, removing mocked transport layers only after live flows verify the contract.

Node.js and the runner scripts inside the plugin root are required for execution. Native macOS applications require running Swift suites in a native macOS CI job, and UI verification requires inspecting actual journey screenshots.

What it helps with

  • bootstrap an end-to-end QA harness for a project
  • generate .agents/qa/config.json and the suite.mjs test runner
  • scan project routes and manifests using project-qa discover
  • wire native macOS Swift suites into CI test pipelines

Example prompts for the agent

  • Set up QA for this backend and web interface
  • Create real integration tests for the browser extension
  • Add project QA journeys for the macOS application

How to install setup

1Plugin marketplace

/plugin marketplace add fcakyon/claude-codex-settings
/plugin install qa@claude-settings

This installs the whole qa plugin with all its skills

2Skills CLI (Vercel)

npx skills add fcakyon/claude-codex-settings --skill setup -a claude-code

3Manually: put the files in place

~/.claude/skills/setup/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/setup/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about setup

How to install setup in Claude Code?

Run /plugin marketplace add fcakyon/claude-codex-settings followed by /plugin install qa@claude-settings. You can also install it directly via npx skills add fcakyon/claude-codex-settings --skill setup -a claude-code.

What does the setup skill do?

It scans repositories using project-qa, locates untested routes, and generates .agents/qa configuration files. Once generated, it helps convert blocked template cases into working integration journeys.

What is required for the setup skill to run?

It requires running npm ci --prefix <plugin-root> and accessing node <plugin-root>/scripts/qa.mjs directly or via npm link. When running CI as root over external checkouts, the path must be added to git's safe.directory.

Source text

SKILL.md · Apache-2.0
---
name: setup
description: This skill should be used when user asks to "set up QA", "create real integration tests", or "add project QA journeys" for websites, extensions, Android apps, native macOS apps or backends.
---

# Set up project QA

The plugin supplies the runner. The project owns journeys, selectors, fixtures and outcome assertions. Execution belongs in tracked scripts. Skills guide discovery and interpretation.

1. Run `npm ci --prefix <plugin-root>` once. Use `node <plugin-root>/scripts/qa.mjs` directly, or `npm link` from the plugin directory to install `project-qa`. In CI, pin the plugin to a commit and bump it on purpose. When CI runs as root over a checkout owned by another user, add it to git's `safe.directory`, or reports record an empty revision and a clean tree.
2. Run `project-qa discover --root <repo>`. Read the owning routes, callers, test commands, manifests and related repositories. Discovery identifies candidate coverage, not working features.
3. Run `project-qa setup --root <repo> --base-url <verified-url>`, adding `--extension <built-directory>` and repeated `--related <relative-repo>` where relevant. It creates `.agents/qa/config.json`, `.agents/qa/suite.mjs` and `.agents/skills/qa/SKILL.md`. Existing files are preserved. Discovery evidence stays in command output, so runtime configuration does not become a stale copy of package manifests.
4. Replace generated blocked cases with real journeys using [the API](../../references/runner.md) and [surface requirements](../../references/surfaces.md). Reuse existing integration tests before writing another owner for the same contract.
5. Configure [sessions](../../references/sessions.md) and [telemetry isolation](../../references/telemetry.md) before live runs. Test accounts have ordinary product permissions.
6. For a native macOS project, wire its real sandboxed Swift owner suite and app self-test into a macOS CI job. Reuse the same command locally and in CI through the shared macOS test CLI or `qa.swiftTests`. Require process success, positive test counts and no skipped tests. Linux adapter checks cannot replace this job. Verify the job is a required release check and executes on the candidate commit.
7. Run the suite, inspect failures, repair incorrect test assumptions, and rerun affected journeys. Run setup again to verify it preserves specialization. Update the local skill with commands and observed coverage. For a repaired bug, follow the regression and CI-selection guidance in [surface requirements](../../references/surfaces.md).

Retain pure algorithm tests with meaningful edge cases. Remove mocked transport tests only after real journeys prove the same contract. A mocked expired-session response does not prove a provider's actual rejection.

A root-page response, discovered component or assertion-free test is not full coverage. Missing credentials, unavailable titles, MFA and absent devices are blocked coverage. Expected outcomes come from independent fixtures and user contracts, not the implementation being tested.

For native macOS permission onboarding, read [the native workflow](../../references/macos.md) and complete its capture-readiness steps before GUI work. The agent must inspect real journey screenshots before claiming UI acceptance. Ask the user only for an OS action they must perform. The shared driver supplies exact AX window selection, native mouse input and scoped captures. Projects own selectors and app outcomes. Run nonmutating protocol and signed-fixture checks in CI, with installed grants only in a separately authorized manual-live run.

© Fatih C. Akyon · Apache-2.0 · fcakyon/claude-codex-settings

Similar skills

All skills for Testing and QA →

setup by other authors comes first.

setup official anthropics setup is a Claude Code skill that checks connected tools, learns your writing style from sent email, and builds a starter dashboard without a questionnaire. For first run. CX Codex +1 371installs setup Alireza Rezvani setup sets up an autoresearch experiment by collecting its target file, evaluation command, metric, and optimization direction. Use it before starting an optimization loop. CX Codex +1 1,480installs setup aws This skill kicks off the connection setup for AWS DevOps Agent and AWS Security Agent. Reach for it when MCP tools go missing or you are wiring the toolkit up for the first time. CX Codex +1 1,600installs setup marketcalls setup sets up a Python backtesting environment with VectorBT and OpenAlgo. It configures the virtual environment, installs dependencies, and prepares .env. CX Codex +1 2,100installs setup assistant-ui setup is a Claude Code skill that installs and configures assistant-ui through its CLI. It covers new projects, existing Next.js apps, and package updates. CX Codex +1 6,400installs om-integration-tests open-mercato om-integration-tests is a Claude Code skill that generates and debugs E2E tests by exploring live applications, reusing repository test runners and shared test environments. CX Codex +1 3,800installs testing sangrokjung testing enforces a strict TDD workflow in Claude Code with an 80% coverage floor, Playwright E2E suites, and three test tiers. CX Codex +1 ★ 880repo stars playwright-interactive-sandbox customware-ai The playwright-interactive-sandbox skill runs iterative web UI checks through a persistent Node.js Playwright workflow. Use it for functional checks, visual QA, and screenshot evidence. CX Codex +1 401installs qa-project-context kindlmann Scans your repo and generates .agents/qa-project-context.md with stack, test tools, and risk areas. Useful when configuring QA context before running other QA skills. CX Codex +1 1,800installs ai-bug-triage kindlmann ai-bug-triage builds a hybrid bug triage pipeline: normalizes CI logs, deduplicates failures via deterministic fingerprints, and uses an LLM to evaluate severity and draft tickets. CX Codex +1 1,700installs shiplight ShiplightAI A router skill for Shiplight testing and QA. It takes a subcommand and hands it to the matching workflow, and only fires when you name shiplight. CX Codex +1 539installs

From the same set113

The whole set →
ui-measure-views ui-measure-views guides agents to measure React Native view dimensions using useLayoutEffect alongside onLayout without triggering unneeded re-renders. CX Codex +1 ★ 1,216repo stars create-issue create-issue is a Claude Code skill for drafting or revising issues about bugs and missing features. It can also refine an issue title or body before publication. CX Codex +1 ★ 1,216repo stars list-performance-item-memo list-performance-item-memo — a rule for React list rows: split the item into primitive props so memo() can compare them shallowly and skip a render when nothing changed. CX Codex +1 ★ 1,216repo stars react-state-dispatcher react-state-dispatcher enforces functional dispatch updaters in React setState calls whenever state depends on current values. Prevents stale closures in callbacks. CX Codex +1 ★ 1,216repo stars list-performance-inline-objects list-performance-inline-objects — a rule that prevents creating objects inside renderItem for props. Inline objects create new references on every render and break memoization. CX Codex +1 ★ 1,216repo stars setup setup configures Tavily MCP server credentials inside Claude Code and saves them to the plugin configuration. CX Codex +1 ★ 1,216repo stars bundle-analyzable-paths The bundle-analyzable-paths rule recommends making import and file paths statically analyzable. That keeps bundlers from treating a broad set of files as reachable. CX Codex +1 ★ 1,216repo stars list-performance-item-expensive The list-performance-item-expensive rule bans heavy list rows: data queries, extra hooks and React Context reads inside renderItem. CX Codex +1 ★ 1,216repo stars

Project sponsors

Companies that keep this analytics open