@venelx/mcp
Enables AI assistants to manage and operate Venelx CI/CD pipelines through the Venelx REST API, including listing projects, triggering builds, checking status, tailing logs, accessing artifacts and workers, and diagnosing signing or GitHub access.
README
@venelx/mcp
An MCP (Model Context Protocol) server that lets AI assistants (Claude Code, Cursor, Kimi Code, and any MCP-compatible client) operate the Venelx CI/CD platform through its REST API: list projects, trigger builds, check build status, tail logs, list artifacts and workers, and diagnose signing / GitHub access.
The server talks to the Venelx API over HTTPS and to your AI client over stdio.
Prerequisites
- Node.js ≥ 18
- A Venelx account with at least one project
- A personal API token — create one in the Venelx dashboard at Account → API tokens: https://app.venelx.com/account/tokens
Token scopes
| Scope | What it allows |
|---|---|
read |
All read-only tools (list projects, status, logs, artifacts, …) |
write |
Everything read allows, plus trigger_build (build-trigger endpoints) |
Tokens look like vx_.... Treat them like passwords — pass them via environment
variables, never commit them.
Configuration
| Environment variable | Required | Default | Description |
|---|---|---|---|
VENELX_TOKEN |
yes | — | Personal API token (vx_...) |
VENELX_API_URL |
no | https://api.venelx.com |
API base URL (self-hosted etc.) |
Install & run
Claude Code
claude mcp add venelx --env VENELX_TOKEN=vx_... -- npx -y @venelx/mcp
Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"venelx": {
"command": "npx",
"args": ["-y", "@venelx/mcp"],
"env": {
"VENELX_TOKEN": "vx_..."
}
}
}
}
Kimi Code / generic MCP client
Same shape as above — add to your client's MCP JSON config:
{
"mcpServers": {
"venelx": {
"command": "npx",
"args": ["-y", "@venelx/mcp"],
"env": {
"VENELX_TOKEN": "vx_...",
"VENELX_API_URL": "https://api.venelx.com"
}
}
}
}
Local development
git clone <repo> && cd venelx-mcp
npm install
npm run build
Then point your MCP client config at the local build:
{
"mcpServers": {
"venelx": {
"command": "node",
"args": ["/path/to/venelx-mcp/dist/index.js"],
"env": {
"VENELX_TOKEN": "vx_..."
}
}
}
}
Tools
| Tool | Scope needed | What it does |
|---|---|---|
list_projects |
read | List projects the token can access (id, name, GitHub URL, platforms, role). |
get_project |
read | Project details plus a GitHub access summary. |
trigger_build |
write | Queue a build for a platform (ios, android, …). Surfaces skipReason (e.g. signing_not_ready, github_token_invalid). |
build_status |
read | Current pipeline status per platform + the 10 most recent builds. |
tail_logs |
read | Last N lines of a build log (latest build, or a given buildId). |
list_artifacts |
read | Build artifacts for a project. |
list_workers |
read | Build workers owned by the account (name, status, capabilities). |
signing_status |
read | Code-signing readiness, optionally per platform (ios/android). |
github_access |
read | GitHub repo access diagnosis: githubAccess {ok, reason, message} + auth source label. |
Troubleshooting
- 401 Authentication failed —
VENELX_TOKENis missing, wrong, or revoked. Create a new token at Account → API tokens (https://app.venelx.com/account/tokens) and update your MCP config. - 403 Forbidden on
trigger_build— your token only has thereadscope. Create a token with the write scope. - Could not reach the Venelx API (ECONNREFUSED / ENOTFOUND) — the API base URL is
unreachable. Check
VENELX_API_URL(defaults tohttps://api.venelx.com) and your network. - Nothing happens / no output — MCP servers speak JSON-RPC on stdout; run it through your MCP client, not directly in a terminal. Diagnostics are printed to stderr.
License
MIT
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。