Provides tools for interacting with Siemens PLCs, enabling reading/writing data blocks, memory, timers, counters, and controlling PLC states.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-snap7 read DB1 from PLC 192.168.0.1"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for python-snap7, enabling MCP clients to interact with Siemens PLCs
pip install mcp-snap7Related MCP server: OpenTIA-MCP
python-snap7 requires the libsnap7 library to be installed on your system. See python-snap7 documentation for installation instructions.
Configure your MCP client to use mcp-snap7 as a stdio server:
{
"mcpServers": {
"mcp-snap7": {
"command": "mcp-snap7"
}
}
}connect_plc - Connect to a Siemens PLC
disconnect_plc - Disconnect from PLC
get_connected - Check connection status
db_read / db_write - Read/write data blocks
mb_read / mb_write - Read/write memory bytes
tm_read / tm_write - Read/write timers
ct_read / ct_write - Read/write counters
eb_read / eb_write - Read/write edge inputs
ab_read / ab_write - Read/write absolute bytes
get_cpu_info - Get CPU information
get_cpu_state - Get CPU state
get_protection - Get PLC protection level
plc_cold_start - Trigger cold start
plc_hot_start - Trigger hot start
plc_stop - Stop PLC
get_error_text - Get error description
git clone https://github.com/daedalus/mcp-snap7.git
cd mcp-snap7
pip install -e ".[test]"
# run tests
pytest
# format
ruff format src/ tests/
# lint
ruff check src/ tests/
# type check
mypy src/Read absolute bytes from PLC.
Inputs: start (required), size (required). Returns: result
Write absolute bytes to PLC.
Inputs: start (required), data (required). Returns: result
Connect to a Siemens PLC.
Inputs: address (required), rack, slot, tcp_port. Returns: result
Read counters from PLC.
Inputs: start (required), size (required). Returns: result
Write counters to PLC.
Inputs: start (required), data (required). Returns: result
Read data block from PLC.
Inputs: db_number (required), start (required), size (required). Returns: result
Write data block to PLC.
Inputs: db_number (required), start (required), data (required). Returns: result
Disconnect from the connected PLC.
Returns: Disconnection status message
Inputs: no named parameters. Returns: result
Read edge inputs from PLC.
Inputs: start (required), size (required). Returns: result
Write edge inputs to PLC.
Inputs: start (required), data (required). Returns: result
Check if PLC is connected.
Returns: True if connected, False otherwise
Inputs: no named parameters. Returns: result
Get CPU information from PLC.
Returns: Dictionary containing CPU information
Inputs: no named parameters. Returns: object
Get CPU state from PLC.
Returns: CPU state string
Inputs: no named parameters. Returns: result
Get error description from error code.
Inputs: error_code (required). Returns: result
Get PLC protection level.
Returns: Dictionary containing protection level info
Inputs: no named parameters. Returns: object
Read memory bytes from PLC.
Inputs: start (required), size (required). Returns: result
Write memory bytes to PLC.
Inputs: start (required), data (required). Returns: result
Trigger PLC cold start.
Returns: Status message
Inputs: no named parameters. Returns: result
Trigger PLC hot start.
Returns: Status message
Inputs: no named parameters. Returns: result
Stop PLC.
Returns: Status message
Inputs: no named parameters. Returns: result
Read timers from PLC.
Inputs: start (required), size (required). Returns: result
Write timers to PLC.
Inputs: start (required), data (required). Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 22 tools
Each tool has a distinct purpose: connecting/disconnecting, reading/writing specific memory areas, retrieving CPU information, and controlling PLC state. There is no overlap or ambiguity.
All tool names follow a consistent verb_noun pattern with snake_case (e.g., connect_plc, db_read, plc_stop). The naming is predictable and easy to parse.
22 tools is on the higher side but appropriate for the complex domain of PLC interaction, covering multiple memory types and control operations. It feels well-scoped without being bloated.
The tool set provides comprehensive read/write access to common PLC memory areas and basic control functions. Minor gaps like safety operations or additional diagnostics exist, but core workflows are covered.
Connect any MCP client to MetaTrader 4/5 to read prices, manage positions, and place trades.
Hosted MCP server connecting AI assistants to 9,000+ apps and 40,000+ actions via Zapier.
MCP gateway and cloud hosting for MCP servers, agents and apps like RAG, managed from your agent.
Persistent memory, knowledge graph, task management and more for AI agents via MCP.
