MCP.so
ログイン

VibeShift Web tester

@Ilikepizza2

リポジトリホームページ

VibeShift Web tester について

[MCP Server] The Security Agent for AI assisted coding

#testing#debugging#qa#web#mcp#debugging-tool#playwright#mcp-server#vibe-coding#vibe-testing

基本情報

カテゴリ

その他

ライセンス

Apache-2.0

ランタイム

python

トランスポート

stdio

公開者

Ilikepizza2

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "VibeShift": {
      "command": "python",
      "args": [
        "-m",
        "venv",
        "venv"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is VibeShift Web tester?

VibeShift Web tester is an MCP server that acts as an automated security engineer for AI coding assistants. It analyzes AI-generated code for vulnerabilities, enables AI-driven remediation, and provides automated browser testing capabilities, including test recording, execution, and discovery.

How to use VibeShift Web tester?

Install the repository, set up a Python virtual environment, install dependencies and Playwright browsers, configure an LLM API key in .env, and add the MCP server configuration to your AI coding assistant. Interact via natural language commands such as “Scan this code for security issues” or “Record a test for login.”

Key features of VibeShift Web tester

  • MCP integration with Cursor, Copilot, Roo Code and others
  • Automated static (Semgrep) and dynamic (Nuclei) security scanning
  • AI-assisted test recording via natural language descriptions
  • Deterministic test execution with Playwright
  • AI-powered test discovery by crawling websites
  • Regression testing and self-healing of existing tests

Use cases of VibeShift Web tester

  • Automated security review of AI-generated code before commit
  • Recording and running browser tests for web applications
  • Discovering and suggesting test steps for websites via crawling
  • Regression testing to catch bugs after code changes

FAQ from VibeShift Web tester

What does VibeShift Web tester do that security scanners alone don’t?

It integrates directly into AI coding workflows, triggering on code generation and returning vulnerability reports that the AI assistant can use to suggest or apply fixes automatically.

What are the runtime requirements?

Python 3.10+, an LLM API key (Gemini 2.0 Flash recommended), and Playwright browsers installed via patchright install --with-deps.

Where are test results and recordings stored?

Recorded tests, execution results, and discovery outputs are saved as JSON files in the output/ directory.

How does authentication work?

No external authentication is needed; configure an LLM API key in the .env file (e.g., LLM_API_KEY="your_key").

Which transports does the server use?

Communication occurs via the Model Context Protocol (MCP), typically over stdin/stdout or a socket, as specified in the MCP configuration.

コメント

「その他」の他のコンテンツ

FastMCP v2 🚀

jlowin

🚀 The fast, Pythonic way to build MCP servers and clients.

26.0K

MaxKB

1Panel-dev

🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。

22.0K

EverArt

modelcontextprotocol

Model Context Protocol Servers

88.1K

Activepieces

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

23.2K

Awesome Mlops

visenger

A curated list of references for MLOps

14.0K