S6S MCP Server

MCP (Model Context Protocol) server for S6S.ai — the workflow engine that AI agents use. Deploy, manage, and monitor persistent multi-step workflows from any MCP-compatible AI agent.

What it does

S6S turns natural language automation requests into reliable, observable workflows. This MCP server gives your AI agent the ability to:

• Deploy workflows — describe what you want automated, the agent builds and deploys it
• Monitor runs — check status, view step-by-step results, debug failures
• Manage workflows — list, update, delete, cancel, retry

Works with Claude Desktop, Claude Code, Cursor, and any MCP-compatible client.

Quick Start

1. Get an S6S API key

Sign up at s6s.ai, go to API Keys, and create a key.

2. Install

npm install -g @s6s/mcp-server

Or run directly with npx:

npx @s6s/mcp-server

3. Configure your AI agent

Set two environment variables:

| Variable | Required | Description | |---|---|---| | S6S_API_KEY | Yes | Your S6S API key (starts with s6s_) | | S6S_BASE_URL | No | API base URL — defaults to https://s6s.ai. Set this if self-hosting (e.g. https://workflows.mycompany.com). |

Setup by Platform

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "s6s": {
      "command": "npx",
      "args": ["@s6s/mcp-server"],
      "env": {
        "S6S_API_KEY": "s6s_your_key_here",
        "S6S_BASE_URL": "https://s6s.ai"
      }
    }
  }
}

Claude Code

Add to .claude/settings.json in your project or ~/.claude/settings.json globally:

{
  "mcpServers": {
    "s6s": {
      "command": "npx",
      "args": ["@s6s/mcp-server"],
      "env": {
        "S6S_API_KEY": "s6s_your_key_here",
        "S6S_BASE_URL": "https://s6s.ai"
      }
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project:

{
  "mcpServers": {
    "s6s": {
      "command": "npx",
      "args": ["@s6s/mcp-server"],
      "env": {
        "S6S_API_KEY": "s6s_your_key_here",
        "S6S_BASE_URL": "https://s6s.ai"
      }
    }
  }
}

OpenClaw

Add as an MCP plugin in your OpenClaw configuration:

{
  "mcpServers": {
    "s6s": {
      "command": "npx",
      "args": ["@s6s/mcp-server"],
      "env": {
        "S6S_API_KEY": "s6s_your_key_here",
        "S6S_BASE_URL": "https://s6s.ai"
      }
    }
  }
}

Available Tools

| Tool | Description | Annotations | |---|---|---| | s6s_get_catalog | Discover all node types, triggers, expressions, and flow syntax | read-only | | s6s_deploy_workflow | Deploy a workflow from a Recipe JSON — single call, auto-runs | | | s6s_validate_workflow | Pre-flight check: credentials, config, schedule, structure | read-only | | s6s_run_workflow | Manually trigger a run for a deployed workflow | | | s6s_get_run_status | Check run progress and step-by-step results | read-only | | s6s_list_workflows | List all deployed workflows | read-only | | s6s_list_runs | List recent runs with optional status filter | read-only | | s6s_cancel_run | Cancel an active run | destructive | | s6s_retry_run | Retry a failed or cancelled run | | | s6s_delete_workflow | Permanently delete a workflow | destructive | | s6s_update_workflow | Update workflow name or enabled status | | | s6s_list_credentials | List connected service credentials (no secrets) | read-only | | s6s_list_integrations | List all platform integrations S6S can connect to | read-only | | s6s_get_integration | Get full details for a specific integration | read-only |

Tools annotated as read-only won't prompt for confirmation in Claude Desktop. Tools annotated as destructive will always prompt.

Prompts

| Prompt | Description | Arguments | |---|---|---| | create-workflow | Guided workflow creation — step-by-step help from catalog to deploy | description: what the workflow should do | | debug-run | Systematic debugging of a failed or stuck run | run_id: the run ID to debug |

Resources

| URI | Description | |---|---| | s6s://catalog | Full workflow catalog (actions, triggers, expressions, flow patterns). Cached for 5 minutes. |

Example

Tell your AI agent:

"Create a workflow that fetches the Hacker News front page every hour and posts the top story to Slack"

The agent will:

1. Call s6s_get_catalog to learn available node types
2. Build a Recipe with a schedule trigger, HTTP fetch, transform, and Slack post
3. Call s6s_deploy_workflow to deploy and start it
4. S6S runs it persistently — survives restarts, retries on failure, fully observable

Recipe Format (what agents send)

{
  "name": "HN to Slack",
  "trigger": { "type": "schedule", "config": { "cron": "0 * * * *" } },
  "steps": [
    { "ref": "fetch_hn", "action": "http", "config": { "url": "https://hacker-news.firebaseio.com/v0/topstories.json", "method": "GET" } },
    { "ref": "get_top", "action": "http", "config": { "url": "https://hacker-news.firebaseio.com/v0/item/{{fetch_hn.output.body[0]}}.json", "method": "GET" } },
    { "ref": "notify", "action": "slack", "config": { "channel": "#news", "textTemplate": "Top HN: *{{get_top.output.body.title}}*\n{{get_top.output.body.url}}" } }
  ],
  "flow": ["fetch_hn", "get_top", "notify"]
}

Links

• S6S.ai — Sign up and manage workflows
• GitHub — Source code