letstakeawalk/dot_claude/skills/qa
dot_claude/skills/qa/SKILL.md
--- name: qa description: "Run QA on the current project. Auto-detects project type and runs browser testing for frontend projects. Use after implementing features to verify they work end-to-end." allowed-tools: Read, Glob, Grep, Bash(bun run:*), Bash(npm run:*), Bash(curl:*), Bash(lsof:*), Bash(kill:*), mcp__plugin_playwright_playwright__* argument-hint: [url or flow to test] [--fix] --- ## Task Verify the project works end-to-end by running it and testing interactively. Not a code review — t
tools
Updated Apr 23, 2026
$ install --global
skillsauthnpx skillsauth add letstakeawalk/dotfiles dot_claude/skills/qaInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 23, 2026, 12:20 PM38.8s1 file scanned
SKILL.md
- name:
- qa
- description:
- Run QA on the current project. Auto-detects project type and runs browser testing for frontend projects. Use after implementing features to verify they work end-to-end.
- allowed-tools:
- Read, Glob, Grep, Bash(bun run:*), Bash(npm run:*), Bash(curl:*), Bash(lsof:*), Bash(kill:*), mcp__plugin_playwright_playwright__*
- argument-hint:
- [url or flow to test] [--fix]
Task
Verify the project works end-to-end by running it and testing interactively. Not a code review — this is runtime behavior testing.
Project Detection
- Check for project markers:
svelte.config.jsor+page.sveltefiles → SvelteKit (browser QA)next.config.*orapp/page.tsx→ Next.js (browser QA)vite.config.*with no framework → Vite SPA (browser QA)
- Detect package manager:
bun.lock/bun.lockb→ bun,package-lock.json→ npm - If none detected, ask user what to test
Browser QA (Frontend)
Setup
- Detect dev server port:
- Read
vite.config.*forserver.port - Check
package.jsonscripts for--portflag - Default: 5173 (Vite/SvelteKit), 3000 (Next.js)
- Read
- Check if dev server already running:
lsof -i :<port> - If not running:
- Start with detected package manager:
bun run devornpm run dev(background) - Set
started_server = true(track for cleanup) - Poll port with
lsof -i :<port>every 2s until ready (timeout 30s)
- Start with detected package manager:
- Navigate to the app URL
Test Flow
- Take accessibility snapshot (
browser_snapshot) to understand page structure - If
$ARGUMENTSspecifies a flow, test that flow - If no arguments, test the current page:
- Verify page loads without errors
- Check console for errors/warnings (
browser_console_messages) - Check network for failed requests (
browser_network_requests) - Screenshot the page (
browser_take_screenshot) - Test interactive elements: click buttons, fill forms, navigate links
- Verify state changes after interactions
- For each issue found:
- Screenshot the broken state
- Note the console errors
- Identify the likely cause
Common Flows to Test
- Navigation between pages
- Form submission (valid + invalid input)
- Auth flow (login/logout) if present
- Empty states and loading states
- Error handling (what happens when API fails?)
- Responsive layout (resize to mobile: 375px width)
Reporting
### QA Report: [project name]
**Environment:** [type] on port [port] | [base url]
### Tested
1. [flow]: pass | fail
[details if fail — console error, screenshot ref, etc.]
### Console Errors
[list any JS errors or warnings, or "None"]
### Network Issues
[list any failed requests, or "None"]
### Verdict: PASS | FAIL | PARTIAL
[one-line summary]
Fix Mode (--fix)
When $ARGUMENTS contains --fix:
- Run QA as normal
- For each failure, attempt to fix the root cause
- Re-test after fix to verify
- Report what was fixed and what remains
Only fix clear bugs (broken handlers, missing imports, wrong selectors). Leave ambiguous UX issues for user.
Rules
- Don't start a dev server if one is already running on the port
- Always check console and network — many bugs are silent in the UI
- Screenshot before and after interactions — visual evidence
- If dev server fails to start, report the error and stop
- Close browser when done (
browser_close) - Kill dev server only if
started_server = true— never kill a server you didn't start - If testing many pages/endpoints, limit to 10 most important. Ask user to prioritize if more
Related Skills
letstakeawalk/test-audit
development
VerifiedTrustedCommunity
Analyze test coverage gaps — untested public APIs, missing edge cases, uncovered error paths. Spawns the test-analyzer agent.
SKILL.mdUpdated Apr 23, 2026
letstakeawalk/dot_claude/skills/review
tools
VerifiedTrustedCommunity
--- name: review description: "Code review — detects language and spawns the appropriate reviewer agent. Works with diffs, staged changes, files, or directories." allowed-tools: Read, Glob, Grep, Edit, Write, Bash(git diff:*), Bash(git log:*), Bash(git status:*), Bash(git branch:*) argument-hint: [file/dir|staged|lang] [security|performance|idioms|bugs|tests] [--fix] [--deep] --- ## Context - Branch: !`git branch --show-current` - Changed files: !`git diff --name-only HEAD` - Diff stats: !`git
SKILL.mdUpdated Apr 23, 2026
letstakeawalk/remember
testing
VerifiedTrustedCommunity
This skill should be used when the user says 'remember this', 'save this for later', 'note this down', 'always do X', 'never do Y', or wants to persist a decision or preference across sessions.
SKILL.mdUpdated Apr 23, 2026
letstakeawalk/dot_claude/skills/refactor
tools
VerifiedTrustedCommunity
--- name: refactor description: "Identify and apply refactoring opportunities — supports full analysis, cleanup, simplify, structure, naming, and DRY focus modes." allowed-tools: Read, Glob, Grep, Edit, Write, Bash(git diff:*), Bash(git branch:*) argument-hint: [file/dir|staged] [full|cleanup|simplify|structure|naming|dry] --- ## Context - Branch: !`git branch --show-current` - Changed files: !`git diff --name-only HEAD` - Diff stats: !`git diff --stat HEAD` ## Task Analyze code for refactori
SKILL.mdUpdated Apr 23, 2026