BugForge
AI-powered QA defect reporting MCP server that generates Jira-ready defect reports with business context, reproduction steps, and priority from test scenarios and error details.
README
BugForge
BugForge is an open-source MCP server for AI-powered QA defect reporting.
Drop in your test scenario and error details — BugForge generates a complete Jira defect report with business context, reproduction steps, failure classification, and priority. Fully config-driven via prompt.md customize the format, tone, and fields to match your team's standards.
Zero API key required. Works natively inside Claude Desktop
Folder structure
bugForge/
├── index.js
├── bugforge_prompt.md
├── package.json
├── README.md
└── LICENSE
Prerequisites
- Node.js v18 or higher
- Claude.ai account (Free or Pro)
- Claude Desktop app
Working
How Does it work ?
BugForge works inside LLM-connected MCP clients (like Claude Desktop). It assembles a structured prompt from your inputs — the host LLM then generates the final defect report automatically. No API key required.
How do I install it?
git clone https://github.com/ervikasti/bugForge.git
cd bugforge
npm install
How do I connect it to Claude Desktop?
1. Open your Claude Desktop config file:
- **Mac:** `~/Library/Application Support/Claude/claude_desktop_config.json`
- **Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
2. Add BugForge under `mcpServers`:
\```json
{
"mcpServers": {
"bugforge": {
"command": "node",
"args": ["/absolute/path/to/bugForge/index.js"]
}
}
}
\```
3. Restart Claude Desktop
Installation via npm (Recommended)
No cloning needed. Add BugForge directly to your MCP client config:
Using npx (zero install)
{
"mcpServers": {
"bugforge": {
"command": "npx",
"args": ["bugforge-mcp"]
}
}
}
How do I use it ?
Just describe your bug in Claude:
"Generate a defect report — user clicks checkout, payment succeeds but order stays pending"
BugForge will produce a Jira-ready report automatically.
Customize for Your Project
- Edit
bugforge_prompt.mdto change the report format, tone, or rules.
BugForge is built to be adapted — make it yours.
Sample Output
<details> <summary>Click to see a sample BugForge defect report</summary>
Input provided to Claude:
Generate a defect report —
Scenario: User completes payment via gateway, gateway returns SUCCESS
but order remains in PENDING state and UI shows no update.
Error: Payment gateway response: SUCCESS | Order status remains: PENDING
| Payment status not updated in UI | No failure message shown to user
BugForge generated report:
Business Context
This issue occurs at the final stage of the payment flow — after the user
submits payment and the gateway confirms success. The order status update
is a critical post-payment operation. Failure at this step means the system
accepts payment but does not fulfil the transaction, directly breaking the
order management capability.
Preconditions
User must have a valid account, items added to cart, and access to a
payment method accepted by the gateway.
Steps to Reproduce
* Navigate to the checkout page.
* Add items to cart and proceed to payment.
* Enter valid payment details and submit.
* Payment gateway returns a SUCCESS response.
* Verify order status in the UI.
Actual Result
Payment gateway response: SUCCESS. Order status remains: PENDING.
Payment status not updated in UI. No failure message shown to user.
Expected Result
After gateway returns SUCCESS, order status should update to CONFIRMED
and payment status should reflect as PAID in both the UI and system.
Business Impact
User has completed payment but order remains pending — creating a ghost
transaction scenario. Causes customer confusion and potential duplicate payments.
Failure Classification & Priority
* Classification: Functional — Post-payment State Sync Failure
* Priority: High
</details>
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。