Requirement Trace Checker
Enables users to analyze project materials and build traceability from requirements to coverage, identifying gaps.
README
Requirement Trace Checker
Requirement Trace Checker is an independent OpenAI App project for building traceability from explicitly supplied project materials.
It analyzes only text that the user provides. The supported chain is:
Requirement -> Acceptance Criteria -> Implementation Evidence -> Test Evidence -> Coverage Status -> Gap
Project Identity
- App Name:
Requirement Trace Checker - App Slug:
requirement-trace-checker - Version:
1.0.0 - Cloudflare Worker name:
requirement-trace-checker - MCP endpoint:
POST /mcp - Support email:
sidcraigau@gmail.com - Current status: not deployed, not in Developer Mode
Do not commit secrets, tokens, or passwords. This Gate 2 project does not include public deployment URLs.
Architecture
- Cloudflare Workers
- TypeScript
- JSON-RPC MCP route at
POST /mcp - Tool schemas and annotations in
src/schemas.ts - Tool runtime logic in
src/tools.ts - Unified response builders in
src/response.ts - Runtime validation in
src/validation.ts - Reserved future page directory in
src/pages/
Validation And Response Boundary
Input schemas define object shape, field types, required fields, enums, array items, and additionalProperties: false.
Runtime validation handles direct MCP calls that bypass client-side schema checks, including missing required fields, wrong types, null, arrays, objects, numbers, booleans, enum values, and extra fields.
Business validation handles empty strings, whitespace-only text, out-of-scope execution requests, and inputs where no explicit requirement can be identified.
All tool success and error outputs are built through src/response.ts. tools/call always returns structuredContent; content is only a compatible display summary.
Tools
Exactly three tools are exposed:
extract_project_requirementsbuild_requirements_traceabilitycheck_requirements_coverage
Unsupported capabilities include repository access, real code execution, external APIs, deployment approval, delivery approval, OAuth, login, databases, email sending, and any external write operation.
Routes
GET /: minimal accessible HTML page containing the app nameGET /healthPOST /mcp
This project does not use /sse or /{app-slug}/mcp.
Local Commands
npm install
npm run typecheck
npm run test
npm run test:mcp:local
npm run dev
npm run test:mcp:local starts a local Worker, checks HTTP routes and MCP JSON-RPC methods, then stops the process it started. It does not rely on public internet, external APIs, or Developer Mode.
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。