worksnaps-mcp
Enables logging offline time and managing Worksnaps projects/tasks from Cursor chat via natural language.
README
@mhdd_24/worksnaps-mcp
MCP server for Worksnaps time tracking from Cursor chat. Verify your API token, list projects and tasks, and log offline time without opening the Worksnaps web UI.
Same architecture as Timelog / Flyway / Notepad++ / Caffeine / Workspace Build MCP packages.
Full documentation: docs/WIKI.md
How it works (30 seconds)
You (chat)
→ worksnaps-mcp
→ Worksnaps REST API (XML over HTTPS)
→ GET /me.xml | /projects.xml | /tasks.xml
→ POST /projects/{id}/time_entries.xml (offline time)
- Configure
WORKSNAPS_API_TOKEN(Profile & Settings → Web Service API in Worksnaps) - Say "list Worksnaps projects" to pick a project ID
- Say "list tasks for project 1234" to pick a task ID
- Say "log 480 minutes to Worksnaps project 1234 task 56" with an optional comment
Prerequisites
| Requirement | Notes |
|---|---|
| Node.js 18+ | Uses native fetch |
| Worksnaps API token | Profile & Settings → Web Service API → Show my API Token |
| Project + task IDs | Use list_projects / list_tasks, or set defaults in env |
Install
Option A — npm (after publish)
npm install -g @mhdd_24/worksnaps-mcp
Option B — npx
npx @mhdd_24/worksnaps-mcp
Option C — clone and build
git clone https://github.com/Mhdd-24/Worksnaps-MCP.git
cd Worksnaps-MCP
npm install
npm run build
node dist/index.js
Configure Cursor
Edit ~/.cursor/mcp.json:
{
"mcpServers": {
"worksnaps": {
"command": "npx",
"args": ["-y", "@mhdd_24/worksnaps-mcp"],
"env": {
"WORKSNAPS_API_TOKEN": "<your-api-token>",
"WORKSNAPS_DEFAULT_PROJECT_ID": "1234",
"WORKSNAPS_DEFAULT_TASK_ID": "5678"
}
}
}
}
Local development:
"command": "node",
"args": ["C:/path/to/worksnaps-mcp/dist/index.js"]
Reload MCP after saving.
Tools
| Tool | Purpose |
|---|---|
whoami |
Resolve current user from API token (GET /me.xml) |
list_projects |
List accessible projects |
list_tasks |
List tasks in a project |
log_time |
Create offline time entry for yourself |
Chat examples
- "Run whoami on Worksnaps"
- "List Worksnaps projects"
- "List tasks for Worksnaps project 3456"
- "Log 6 hours offline time to Worksnaps project 3456 task 12 — comment: Sprint development"
Environment variables
| Variable | Required | Purpose |
|---|---|---|
WORKSNAPS_API_TOKEN |
Yes* | API token (Basic auth username; password ignored) |
WORKSNAPS_DEFAULT_PROJECT_ID |
No | Default project for log_time |
WORKSNAPS_DEFAULT_TASK_ID |
No | Default task for log_time |
WORKSNAPS_BASE_URL |
No | API base URL (default https://api.worksnaps.com/api) |
*Can be passed as token per tool call instead.
Notes
log_timecreates offline time entries (Worksnaps API: POST/projects/{project_id}/time_entries.xml).from_timestampis aligned to a 10-minute boundary (Worksnaps API requirement).- Default
startHouris 9 (local time) when only a date is provided.
License
ISC
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。