Support Ticket Validator
Validates support ticket data by checking required fields and returning a structured completeness status.
README
Support Ticket Validator
Support Ticket Validator checks whether supplied support ticket data contains required fields and returns a structured completeness status.
Boundary
This app is read-only, stateless, deterministic, and has no external side effects. It does not authenticate users, call external APIs, use OAuth, store data, update support systems, reply to customers, approve refunds, close tickets, assign responsibility, or provide open-ended support strategy.
Developer Mode testing, production deployment, and formal submission are outside the current gate and have not been performed.
Technology
- Cloudflare Workers
- TypeScript
- npm
- wrangler
Project Structure
support-ticket-validator/
├── package.json
├── package-lock.json
├── tsconfig.json
├── wrangler.jsonc
├── README.md
├── src/
│ ├── html.d.ts
│ ├── index.ts
│ ├── schemas.ts
│ ├── tool.ts
│ └── pages/
│ ├── home.html
│ ├── privacy.html
│ ├── terms.html
│ └── support.html
└── scripts/
└── test-mcp-local.mjs
Install
npm install
Local Development
npm run dev
The local Worker defaults to:
http://127.0.0.1:8787
Checks
TypeScript:
npm run typecheck
GET /:
curl http://127.0.0.1:8787/
GET /health:
curl http://127.0.0.1:8787/health
Local MCP regression:
npm run test:mcp:local
To run the regression against another local URL:
BASE_URL=http://127.0.0.1:8787 npm run test:mcp:local
URLs
- Website URL:
https://support-ticket-validator.sidcraigau.workers.dev - MCP URL:
https://support-ticket-validator.sidcraigau.workers.dev/mcp - Privacy URL:
https://support-ticket-validator.sidcraigau.workers.dev/privacy - Terms URL:
https://support-ticket-validator.sidcraigau.workers.dev/terms - Support URL:
https://support-ticket-validator.sidcraigau.workers.dev/support
MCP Tool
- Tool name:
validate_support_ticket_completeness - Tool title:
Validate Support Ticket Completeness
Input fields:
ticket: support ticket fields keyed by field name.required_fields: field names that should be present and non-empty.source_label: optional source label.request_intent: optional intent text used only for runtime boundary checks.
Output fields:
statusis_completechecked_fieldspresent_fieldsmissing_fieldssource_labelerrors
Schema / Runtime Boundary
The JSON Schema layer declares top-level shape, allowed fields, and basic JSON types. It intentionally does not set minProperties, minItems, or string minLength, because empty tickets, empty required field lists, blank field names, and empty business values must reach the Tool Runtime and return the unified Error Contract.
Runtime type validation handles non-object arguments, missing required inputs, non-object tickets, array or null tickets, non-array required_fields, non-string required_fields items, and invalid optional string fields.
Business validation handles empty ticket data, empty normalized required field lists, duplicate and whitespace field names, empty values, and out-of-scope request intents.
Error Contract
All success, incomplete, and error responses use the same outputSchema. Error records contain:
code: one ofmissing_required_input,invalid_input_type,empty_input,out_of_scope,internal_errormessage: human-readable error summaryfield: input field related to the error, or an empty string for general errors
Prohibited Capabilities
This project does not include customer replies, refund approval, refund rejection, ticket closure, ticket status updates, classification, extraction from unstructured messages, CRM or support platform integration, database access, OAuth, login, AI model calls, external APIs, emails, or external writes.
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。