Skip to main content
The pflow mcp command group manages Model Context Protocol (MCP) server connections. MCP servers expose external tools that can be used in pflow workflows.
Setup commands. You run add, remove, and sync to configure MCP servers. Your agent uses tools and info to discover available tools when building workflows.

Commands

pflow mcp add

Add MCP servers from config files or JSON, then discover their tools. A server that can’t be reached is still added; pflow prints its error and the pflow mcp sync <name> command to retry.
Arguments:
  • CONFIG_SOURCE - One or more config file paths or raw JSON strings
Examples:
Config formats: Simple format (recommended):
Full MCP format (compatible with Claude Desktop):
HTTP server format:
Environment variables in env and headers support ${VAR} and ${VAR:-default} syntax. They expand at runtime.

pflow mcp list

List all configured MCP servers.
Options:
  • --json - Output as JSON
Output:

pflow mcp remove

Remove an MCP server configuration.
Arguments:
  • NAME - Server name to remove
Options:
  • --force, -f - Skip confirmation prompt
Examples:
Removing a server also removes all its tools from the registry.

pflow mcp sync

Discover and register tools from MCP servers.
Arguments:
  • NAME - Server name to sync (conflicts with --all)
Options:
  • --all, -a - Sync all configured servers
Examples:
Output:
pflow mcp add syncs the servers it adds, and workflow runs never sync. Run pflow mcp sync after editing mcp-servers.json by hand, or to retry a server that failed during add.

pflow mcp list

List registered MCP tools.
Arguments:
  • SERVER - Filter by server name (optional)
Options:
  • --json - Output as JSON
Examples:
Output:

pflow mcp describe

Show detailed information about an MCP tool.
Arguments:
  • TOOL - Tool name to inspect
Example:
Output:
An object parameter, or a list of objects, lists its fields indented beneath it. The usage snippet sets only the required parameters; the rest keep their defaults. Run pflow mcp sync --all after upgrading pflow so tools registered earlier pick up the current parameter types.

pflow mcp serve

Run pflow as an MCP server for AI tools.
Options:
  • --debug - Enable debug logging
This command starts pflow as an MCP server using stdio transport. AI tools connect to pflow and can use its workflow capabilities.
This is typically invoked by AI tools automatically, not run directly. See the integrations guide for setup instructions.
Exposed tools: The MCP server exposes tools for:
  • Discovering workflows and nodes
  • Running workflows
  • Validating workflows
  • Saving workflows to the library
  • Managing settings

Tool naming convention

MCP tools are registered with the pattern mcp-{server}-{tool}: Agents can use the shortest unambiguous name in workflows.

Using MCP tools in workflows

After syncing, MCP tools work like any other node:

File locations

Common workflows

Initial setup

Troubleshooting