Atlas / MCP servers / lbnl-eta / EnergyPlus

EnergyPlusSAFE

Name: EnergyPlus
Author: lbnl-eta

mcp/lbnl-eta/energyplus

MCP server by lbnl-eta · Audited 2026-10-07

The first open-source Model Context Protocol server enabling AI assistants and applications to interact programmatically with EnergyPlus building energy simulation.

Verdict

SAFE

Grade

Trust score

89 /100

Exposed tools

35 23r · 11w · 1d

Transport

stdio · streamable-http

License

NOASSERTION

Stars

120

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

A Model Context Protocol (MCP) server that provides 35 comprehensive tools for working with EnergyPlus building energy simulation models. This server enables AI assistants and other MCP clients to load, validate, modify, and analyze EnergyPlus IDF files through a standardized interface.

Version: 0.1.0 EnergyPlus Compatibility: 26.1.0 (default; see Building against a different EnergyPlus version) Python: 3.10+

📑 Table of Contents

Overview
Installation
Using the MCP Server
Claude Desktop
VS Code
Cursor
Development Setup
VS Code Dev Container
Docker Setup
Local Development
Streamable HTTP Transport (Local Testing)
Building against a different EnergyPlus version
Available Tools
Usage Examples
Architecture
Configuration
Troubleshooting
Contributing
Cite this work
License

Overview

EnergyPlus MCP Server makes EnergyPlus building energy simulation accessible to AI assistants and automation tools through the Model Context Protocol.

Key Features:

🏗️ Complete Model Lifecycle: Load, validate, analyze, modify, and simulate IDF files
🔍 Deep Building Analysis: Extract detailed information about zones, surfaces, materials, and schedules
🚀 Automated Simulation: Execute EnergyPlus simulations with weather files
📊 Advanced Visualization: Create interactive plots and HVAC system diagrams
🔧 **HVAC Intelli

Read from source at commit c65ffbe053deOBSERVED · 2026-10-07

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code

claude mcp add energyplus-mcp-server --env MCP_TOKENS=${MCP_TOKENS} -- uvx energyplus-mcp-server

claude-desktop

{
  "mcpServers": {
    "energyplus-mcp-server": {
      "command": "uvx",
      "args": [
        "energyplus-mcp-server"
      ],
      "env": {
        "MCP_TOKENS": "${MCP_TOKENS}"
      }
    }
  }
}

Exposed tools (35)

23 read · 11 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`add_coating_outside`	write
`add_output_meters`	write
`add_output_variables`	write
`add_window_film_outside`	write
`change_infiltration_by_mult`	read
`check_simulation_settings`	read
`clear_logs`	destructive
`copy_file`	read
`create_interactive_plot`	write
`discover_hvac_loops`	read
`get_error_logs`	read
`get_loop_topology`	read
`get_materials`	read
`get_model_summary`	read
`get_output_meters`	read
`get_output_variables`	read
`get_server_configuration`	read
`get_server_logs`	read
`get_server_status`	read
`get_surfaces`	read
`inspect_electric_equipment`	read
`inspect_lights`	read
`inspect_people`	read
`inspect_schedules`	read
`list_available_files`	read
`list_zones`	read
`load_idf_model`	read
`modify_electric_equipment`	write
`modify_lights`	write
`modify_people`	write
`modify_run_period`	write
`modify_simulation_control`	write
`run_energyplus_simulation`	write
`validate_idf`	read
`visualize_loop_diagram`	read

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

Layer	What it checks	Result
L0	Provenance & inventory	PASS
L1	Static analysis of the code	PASS
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (6 observation(s))
Network: declared (3 observation(s))
Shell: none-observed
Dependencies: pinned
Secrets in source: none-found

Findings (7)

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59

clear_logs

Why it matters. 1 tool(s) can delete or overwrite

Fix. prefer a read-only mode or scoped tokens; the page states the blast radius

LOWNetwork egress · net.raw_ip · CWE-200, CWE-319

README.md:300

Either way you should see a log line like `Listening on http://0.0.0.0:8000 (path=/mcp, 1 tokens)`.