AI-powered E2E testing MCP server. Detects and auto-fixes UI bugs via DevQA Loop and Vision AI.
Config is the same across clients — only the file and path differ.
{
"mcpServers": {
"io-github-ksgisang-awt": {
"args": [
"-y",
"skills"
],
"command": "npx"
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
Your AI coding tool is smart. But it can't see or click a web page. AWT gives it a real browser — so it can test, find bugs, and fix them without you lifting a finger.
Run this in your terminal to verify the server starts. Then let us know if it worked — your result helps other developers.
npx -y 'skills' 2>&1 | head -1 && echo "✓ Server started successfully"
After testing, let us know if it worked:
Five weighted categories — click any category to see the underlying evidence.
No known CVEs.
Checked skills against OSV.dev.
Click any tool to inspect its schema.
Be the first to review
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Others in developer-tools / browser
Read, write, and manage files on the local filesystem
Browser automation with Puppeteer for web scraping and testing
A Model Context Protocol (MCP) server and CLI that provides tools for agent use when working on iOS and macOS projects.
Manage Supabase projects — databases, auth, storage, and edge functions
MCP Security Weekly
Get CVE alerts and security updates for io.github.ksgisang/awt and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
Your AI coding tool is smart. But it can't see or click a web page. AWT gives it a real browser — so it can test, find bugs, and fix them without you lifting a finger.
Works without AI API — write YAML scenarios manually and run
aat run. No API key needed. AI coding tools make it easier, but are not required.
AWT is the execution engine for AI-powered E2E testing. Your AI coding tool designs the tests; AWT runs them in a real browser with Playwright.
How it works in Skill Mode (no extra AI API key needed):
You: "Test the login flow on https://mysite.com"
Your AI coding tool:
→ Writes 5 YAML scenarios (30 steps)
→ Runs: aat run scenarios/
→ Reads failure: "Step 4: Dashboard text not visible"
→ Reads screenshot + source code
→ Fixes src/pages/login.js:23
→ Re-runs: aat run scenarios/
→ All 5 scenarios pass ✓
Key capabilities:
aat loop auto-fixes failures with its own AInpx skills add ksgisang/awt-skill --skill awt -g
git clone https://github.com/ksgisang/awt-skill.git /tmp/awt-skill
cp -r /tmp/awt-skill/awt ~/.claude/skills/awt
rm -rf /tmp/awt-skill
git clone https://github.com/ksgisang/awt-skill.git /tmp/awt-skill
cp -r /tmp/awt-skill/awt .claude/skills/awt
rm -rf /tmp/awt-skill
After installation, the skill auto-triggers on testing-related prompts.
# Re-run the same install command to get the latest version
npx skills add ksgisang/awt-skill --skill awt -g -y
For manual installations: cd ~/.claude/skills/awt && git pull
| Feature | webapp-testing | playwright-skill | AWT |
|---|---|---|---|
| Scenario format | Python scripts | Python scripts | Declarative YAML |
| No-code test creation | No | No | Yes |
| Natural language → test | No | No | Yes |
| Self-healing loop | No | No | Yes (DevQA Loop) |
| Auto-fix on failure | No | No | Yes (AI → PR) |
| Visual matching (OpenCV + OCR) | No | No | Yes (OpenCV + OCR) |
| Pattern learning | No | No | Yes (SQLite) |
| Server lifecycle mgmt | Yes | Yes | Yes (aat dashboard) |
| Multiple AI providers | No | No | Yes (5 providers) |
| Human-like interaction | No | No | Yes (Bezier mouse) |
| Approval modes | No | No | Yes (manual/branch/auto) |
| Cost optimization | N/A | N/A | $0.02–0.05/test |
| Live screenshots | No | Screenshot only | Yes (WebSocket stream) |
| Multi-language OCR | No | No | Yes (10+ languages) |
| Offline support | No | No | Yes (Ollama) |
| Cost tracking | No | No | Yes (per-call logging + aat cost) |
| Source code analysis | No | No | Yes (Skill-exclusive) |
| Scenario caching | No | No | Yes (same spec = no re-call) |
| Browser test overlay | No | No | Yes (live step progress) |
| Dependency ordering | No | No | Yes (depends_on field) |
| Skill Mode (no extra AI cost) | No | No | Yes |
| Canvas/Flutter OCR fallback | No | No | Yes (auto) |
| Platform auto-detection | No | No | Yes (7 frameworks) |
| Structured failure diagnosis | No | No | Yes (AI-independent) |
| Multi-document YAML | No | No | Yes (--- separator) |
| Strict validation | No | No | **Yes (aat va |