Laminar MCP Server
Integrates Laminar workspaces with AI editors to manage workflows, steps, and configuration stores. It allows users to execute, monitor, and debug workflow executions through natural language commands.
README
Laminar MCP Server
Bring your Laminar workspace into Cursor and Claude Code. Read executions, search and debug them, edit workflows, manage configurations, and more — all from your AI-powered editor.
Features
- Workspaces — list, inspect, manage workspace members
- Workflows — create, update, clone, delete, restore workflows
- Flows (Steps) — read/write individual steps, view version history, bulk update
- Executions — list, search, filter by date/status, inspect full payloads, view per-step results
- Execute — trigger workflows synchronously or async, poll for status
- Configuration Stores — manage key-value configs used by
{{config.xxx}}references - Issues — create and manage workspace issues
- Stats — flow execution statistics, recent runs
- Prompts — built-in Laminar workflow specification guide and execution debugger
Quick Start
cd mcp
npm install
npm run setup
This opens a browser window where you sign in with your Laminar account. That's it — tokens are stored at ~/.laminar/tokens.json and the Cursor MCP config is written automatically. Tokens auto-refresh in the background.
Restart Cursor to activate.
Manual Setup
If you prefer env vars over the browser flow:
npm install && npm run build
Set LAMINAR_API_KEY in your Cursor MCP config:
{
"mcpServers": {
"laminar": {
"command": "node",
"args": ["/absolute/path/to/laminar/mcp/dist/index.js"],
"env": {
"LAMINAR_API_KEY": "your-api-key-here"
}
}
}
}
Auth priority
LAMINAR_API_KEYenv varLAMINAR_ACCESS_TOKENenv var- Stored tokens from
npm run setup(~/.laminar/tokens.json) — auto-refreshed
Claude Code
claude mcp add laminar node /absolute/path/to/laminar/mcp/dist/index.js \
-e LAMINAR_API_KEY=your-api-key-here
Available Tools
Workspace & Users
| Tool | Description |
|---|---|
list_workspaces |
List all workspaces |
get_workspace |
Get workspace details |
get_workspace_users |
List workspace members |
get_current_user |
Get authenticated user info |
Workflows
| Tool | Description |
|---|---|
list_workflows |
List workflows in a workspace |
list_archived_workflows |
List archived/deleted workflows |
get_workflow |
Get workflow details |
create_workflow |
Create a new workflow |
update_workflow |
Update name/description |
delete_workflow |
Archive a workflow |
restore_workflow |
Restore archived workflow |
clone_workflow |
Clone a workflow |
Flows (Steps)
| Tool | Description |
|---|---|
list_workflow_flows |
List all steps in a workflow |
get_flow |
Get step details |
read_flow |
Read step program code |
create_flow |
Add a new step |
update_flow |
Update step code/metadata |
delete_flow |
Remove a step |
create_or_update_flows |
Bulk create/update steps |
get_flow_versions |
View step version history |
read_flow_version |
Read historical step code |
Executions
| Tool | Description |
|---|---|
list_executions |
Search/filter executions (date, status, text) |
get_execution |
Get execution details with all step results |
get_execution_status |
Quick status poll (for async) |
get_execution_result |
Get final step output only |
get_full_execution |
Get untruncated execution data |
get_global_workflow_object |
Get shared execution state |
get_flow_run_response |
Get a step's response data |
get_flow_run_transformation |
Get a step's input data |
get_flow_run_program |
Get the code that ran for a step |
Execute Workflows
| Tool | Description |
|---|---|
execute_workflow |
Run workflow synchronously |
execute_workflow_async |
Trigger async execution |
Configuration Stores
| Tool | Description |
|---|---|
list_config_stores |
List configs in workspace |
get_config_store |
Get config store details |
get_config_properties |
Get all key-value pairs |
get_config_property |
Get single property value |
update_config_property |
Set a property |
remove_config_property |
Delete a property |
create_config_store |
Create new config store |
delete_config_store |
Archive config store |
restore_config_store |
Restore archived config |
Issues
| Tool | Description |
|---|---|
list_issues |
List workspace issues |
get_issue |
Get issue details |
create_issue |
Create an issue |
update_issue |
Update issue status/details |
delete_issue |
Delete an issue |
Stats & Misc
| Tool | Description |
|---|---|
get_flow_stats |
Execution statistics |
get_recent_flow_runs |
Recent runs across workflows |
list_api_keys |
List workspace API keys |
Prompts
| Prompt | Description |
|---|---|
laminar-workflow-guide |
Complete Laminar workflow specification — step types, data access, libraries, best practices |
debug-workflow-execution |
Feed a failed execution into the AI for root cause analysis and fix suggestions |
Example Usage in Cursor
"List all workflows in my workspace"
"Show me the last 5 failed executions for workflow 42"
"Read the code for step 3 of workflow 15 and fix the bug that's causing 400 errors"
"Create a new workflow that fetches data from the GitHub API and transforms it"
"What's the execution history for workflow 100 this week?"
"Debug execution 5678 of workflow 42 — why did it fail?"
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。