PLCs.ai — Claude Code MCP server: features and setup
Connect an agent to Allen-Bradley and Siemens PLC projects, tag values, and analysis. Ask about a project or prepare a code change for review.
claude mcp add --transport http plcs-ai https://mcp.plcs.ai/mcpWhat PLCs.ai does
PLCs.ai gives an agent access to PLC projects, current tag values, and project analysis. Its tools can find projects, answer questions with citations, inspect tag history, and flag issues such as dead code or interlock problems. Access to each tool is governed by the permissions attached to an API key or OAuth grant.
Code changes go through a plan and a reviewable proposal; saving a version is a separate action. The server cannot create projects. You need a PLCs.ai account or API key to connect. Using the mcp-remote bridge also requires Node 18+.
What it helps with
- find an organization’s PLC projects and check for live data
- trace interlocks that prevent a conveyor from restarting
- inspect a tag’s recent values
- prepare a code change for review before saving a version
Example prompts for the agent
- List my PLC projects and tell me which ones have live data.
- Trace the interlocks that stop the conveyor from restarting after a stop.
- Draft a proposal to fix the missing handshake between
Conv001andRob010; do not save a version.
How to install PLCs.ai
Platform CLI: remote server
claude mcp add --transport http plcs-ai https://mcp.plcs.ai/mcpPlatform CLI: remote server
codex mcp add plcsai --url https://mcp.plcs.ai/mcpConfig file
{
"mcpServers": {
"plcsai": {
"url": "https://mcp.plcs.ai/mcp"
}
}
}Config file
{
"servers": {
"plcsai": {
"type": "http",
"url": "https://mcp.plcs.ai/mcp"
}
}
}Platform CLI: remote server
gemini mcp add -t http plcsai https://mcp.plcs.ai/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about PLCs.ai
How do I install PLCs.ai in Claude Code?
Run claude mcp add --transport http plcs-ai https://mcp.plcs.ai/mcp, then open /mcp and choose Authenticate to sign in to your PLCs.ai account.
What does PLCs.ai do?
It connects an agent to PLC projects and tag data so the agent can answer questions, inspect analyses, and prepare code proposals. You can review a proposal before saving a version.
Which PLC projects and clients does PLCs.ai support?
The server covers Allen-Bradley and Siemens projects. Listed clients include Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, and Gemini CLI.
