ansys-aedt-mcp
Provides tools for controlling Ansys Electronics Desktop (AEDT) through PyAEDT and native AEDT APIs, enabling AI agents to automate electromagnetic simulations including HFSS, Maxwell, Q3D, Icepak, and Circuit.
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., "@ansys-aedt-mcp Start a new HFSS design and run a frequency sweep"
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.
Ansys AEDT MCP Server
A Model Context Protocol server for Ansys Electronics Desktop, PyAEDT, HFSS, Maxwell, Q3D, Icepak, Circuit, reports, sweeps, and simulation automation.
English · 简体中文 · Documentation · Verification
Why This Exists
ansys-aedt-mcp lets AI agents and MCP clients control Ansys Electronics Desktop through a structured server interface. It combines dedicated tools for common electromagnetic simulation workflows with a broad PyAEDT/native AEDT bridge for advanced automation.
SEO keywords: Ansys MCP, AEDT MCP server, Ansys Electronics Desktop automation, PyAEDT MCP, HFSS automation, HFSS far-field automation, Touchstone export, Touchstone import, signal integrity automation, Maxwell automation, Q3D automation, Icepak automation, AI simulation agent, electromagnetic simulation automation, EDA agent tools, CAE automation.
Related MCP server: COMSOL MCP Server
Highlights
MCP-ready: stdio, SSE, and streamable HTTP transports through the official Python MCP SDK.
PyAEDT-first: HFSS, Maxwell, Q3D/Q2D, Icepak, Circuit, Twin Builder, EMIT, RMxprt, Mechanical, and HFSS 3D Layout entry points.
Native AEDT bridge:
odesktop,oproject,odesign,oeditor, andodesign.GetModule(...)access through controlled tools.Simulation workflow tools: variables, datasets, geometry, materials, setup inspection/update, HPC options, sweeps, optimizations, analysis, reports, field plots, far-field, antenna/RCS data, signal-integrity expressions, Q3D nets, Icepak fan data, Touchstone data/import/export, monitors, and exports.
Agent-safe verification: unit tests run without AEDT licenses; Desktop/native smoke tests run on licensed Windows AEDT installations.
Non-commercial source availability: research, education, personal experimentation, and public knowledge use under PolyForm Noncommercial 1.0.0.
Quick Start
git clone https://github.com/LaplaceYoung/ansys-aedt-mcp.git
cd ansys-aedt-mcp
uv sync --extra dev
uv run ansys-aedt-mcpUse the MCP inspector during development:
uv run mcp dev src/ansysmcp/server.pyRun streamable HTTP:
uv run ansys-aedt-mcp --transport streamable-httpMCP Client Config
{
"mcpServers": {
"ansys-aedt": {
"command": "uv",
"args": ["--directory", "F:\\实验\\ansysmcp", "run", "ansys-aedt-mcp"]
}
}
}After cloning from GitHub, replace the local path with your checkout path.
Tool Surface
Area | Tools |
Environment/API discovery |
|
Session |
|
Project/design |
|
Variables/datasets |
|
Modeling/materials |
|
Ports/sources |
|
Solver-specific controls |
|
Simulation |
|
Exploration |
|
Post-processing |
|
Export |
|
Deletion |
|
Project/design maintenance |
|
Configuration |
|
Native/OO properties |
|
Broad API/workflows |
|
Current MCP registration: 109 tools.
Example Workflows
Desktop/native workflow:
aedt_environment
aedt_start_session(app_name="desktop", version="2024.2", non_graphical=true)
aedt_new_project(project_name="MCPNativeProject")
aedt_insert_design(design_type="HFSS", design_name="HFSSDesign1", solution_type="DrivenModal")
aedt_call(target="omodule", module_name="AnalysisSetup", method="...")PyAEDT solver workflow:
aedt_start_session(app_name="hfss", version="2024.2", non_graphical=true)
aedt_set_variable(name="w", expression="10mm")
aedt_create_geometry(primitive="box", args=[[0, 0, 0], ["w", "5mm", "1mm"]])
aedt_modeler_operation(method="move", args=[["Box1"], ["1mm", "0mm", "0mm"]])
aedt_assign_material(assignment="Box1", material="copper")
aedt_materials_summary
aedt_materials_operation(method="add_material", args=["demo_material"], kwargs={"properties": {"permittivity": 3.2}})
aedt_create_port(method="wave_port", args=["Face1"], kwargs={"name": "P1"})
aedt_hfss_operation(method="create_scattering", kwargs={"ports": ["P1"]})
aedt_create_setup(name="Setup1")
aedt_update_setup(name="Setup1", properties={"MaximumPasses": 8})
aedt_create_frequency_sweep(sweep_kind="linear_count", args=["Setup1", "GHz", 1, 10])
aedt_create_parametric_sweep(variable="w", start="5mm", stop="20mm", step="5mm")
aedt_optimetrics_summary
aedt_optimetrics_setup_operation(collection="parametrics", setup_name="Parametric1", method="update", kwargs={"update_dictionary": {"SaveFields": true}})
aedt_set_hpc_options(cores=8, tasks=2)
aedt_analyze_setup(name="Setup1", cores=8, blocking=true)
aedt_create_output_variable(variable="s11", expression="dB(S(1,1))")
aedt_get_evaluated_value(name="w", units="mm")
aedt_create_report(expressions="dB(S(1,1))")
aedt_post_summary
aedt_post_operation(method="export_report_to_jpg", args=["outputs", "S11"], kwargs={"width": 1200})
aedt_get_traces_for_plot(kwargs={"setup": "Setup1"})
aedt_get_touchstone_data(setup="Setup1")
aedt_signal_integrity_expressions(drivers=["P1"], receivers=["P2"], math_formula="dB")
aedt_insert_far_field(kwargs={"name": "FF1", "theta_step": 5, "phi_step": 5})
aedt_get_antenna_data(setup="Setup1", sphere="FF1")
aedt_get_rcs_data(setup="Setup1", expression="ComplexMonostaticRCSTheta")
aedt_export_report(report_name="S11", output_path="outputs")
aedt_export_touchstone_data(setup="Setup1", sweep="Sweep1", output_file="outputs/design.s2p")
aedt_export_diagnostics(export_kind="convergence", setup="Setup1")AEDT Requirements
Windows with Ansys Electronics Desktop installed for real AEDT execution.
PyAEDT-compatible AEDT version. Local verification detected AEDT
2024.2.Solver-specific licenses for solver app constructors and analysis workflows.
Python 3.10+; the repository pins local development to Python 3.12 through
.python-version.
Verification
uv sync --extra dev
uv run ruff check .
uv run pytest
uv run ansys-aedt-mcp --help
uv run python scripts/aedt_smoke.py --mode environment
uv run python scripts/aedt_smoke.py --mode desktop --version 2024.2 --create-project MCPNativeProjectCurrent local status:
ruff check: passingpytest: 39 passing testsMCP tools registered: 109
Desktop/native AEDT smoke: passing on AEDT 2024 R2
Documentation
Roadmap
Add solver-license-backed end-to-end HFSS smoke examples.
Expand canonical templates for HFSS, Maxwell 3D, Q3D, Icepak, and Circuit workflows.
Add MCP prompts for common AEDT modeling and post-processing tasks.
Add solver-backed smoke examples around diagnostics, near-field extraction, Touchstone analysis, and monitor data.
Add generated API maps for PyAEDT modeler, post, modules, and native AEDT module names.
Community
Contributions are welcome for examples, solver-specific wrappers, docs, tests, and verified workflows. Read CONTRIBUTING.md, SECURITY.md, and CODE_OF_CONDUCT.md.
License
This project uses PolyForm Noncommercial License 1.0.0.
Commercial licensing and redistribution permissions require a separate written agreement.
Available Tools
109 toolsaedt_analyzeC
Run analysis for a named setup or the active design.
Inputs: setup_name, analyze_all. Returns: object
aedt_analyze_setupC
Analyze a named setup with PyAEDT distribution and blocking options.
Inputs: name, cores, tasks, gpus, +6 more. Returns: object
aedt_api_manifestB
List PyAEDT app constructors and method signatures without launching AEDT.
Inputs: app_name, include_private. Returns: object
aedt_apply_solved_variationC
Apply a solved variation dictionary to the active design context.
Inputs: variation (required). Returns: object
aedt_assign_boundary_or_excitationC
Call a public assign_/create_ boundary or excitation method on the active PyAEDT app.
Inputs: method (required), args, kwargs. Returns: object
aedt_assign_materialC
Assign an AEDT material to one object or many objects.
Inputs: assignment (required), material (required). Returns: object
aedt_batch_callB
Execute ordered public PyAEDT/native AEDT calls in one MCP request.
Inputs: operations (required), continue_on_error. Returns: object
aedt_callC
Invoke a public PyAEDT or AEDT native method for broad AEDT API coverage.
Inputs: target (required), attr_path, method, args, +3 more. Returns: object
aedt_change_design_settingsC
Apply active design settings through PyAEDT change_design_settings.
Inputs: settings (required). Returns: object
aedt_change_validation_settingsC
Apply active design validation settings.
Inputs: entity_check_level, ignore_unclassified, skip_intersections. Returns: object
aedt_circuit_operationC
Run an allowlisted Circuit schematic, source, Touchstone, or excitation operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_cleanup_solutionC
Clean solution data through PyAEDT cleanup_solution options.
Inputs: variations, entire_solution, field, mesh, +1 more. Returns: object
aedt_configuration_operationC
Run an allowlisted configuration import, export, or monitor update operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_configuration_summaryA
Return configuration object type and export/import option values.
Inputs: no named parameters. Returns: object
aedt_create_datasetC
Create a 1D or 3D AEDT dataset.
Inputs: name (required), x (required), y (required), z, +7 more. Returns: object
aedt_create_field_plotC
Create a field plot through the active post processor.
Inputs: plot_kind (required), assignment (required), quantity (required), setup, +4 more. Returns: object
aedt_create_frequency_sweepC
Create a frequency sweep, such as linear_count, linear_step, or single_point.
Inputs: sweep_kind (required), args, kwargs. Returns: object
aedt_create_geometryC
Create geometry through app.modeler, for example primitive='box' maps to create_box.
Inputs: primitive (required), args, kwargs. Returns: object
aedt_create_open_regionC
Create an HFSS open region around the active model.
Inputs: frequency, boundary, apply_infinite_ground, gp_axis. Returns: object
aedt_create_optimizationC
Create an Optimetrics optimization, sensitivity, statistical, DOE, or DX setup.
Inputs: calculation, ranges, variables, optimization_type, +8 more. Returns: object
aedt_create_output_variableC
Create a PyAEDT output variable for post-processing expressions.
Inputs: variable (required), expression (required), solution, context. Returns: object
aedt_create_parametric_sweepC
Create an Optimetrics parametric sweep for a variable.
Inputs: variable (required), start (required), stop (required), step (required), +2 more. Returns: object
aedt_create_portC
Create a common HFSS/Circuit port through a controlled PyAEDT method.
Inputs: method (required), args, kwargs. Returns: object
aedt_create_reportC
Create an AEDT report through PyAEDT post-processing.
Inputs: expressions (required), report_category, setup_name, variations, +3 more. Returns: object
aedt_create_setupC
Create an AEDT solution setup and apply setup properties.
Inputs: name, setup_type, properties. Returns: object
aedt_create_touchstone_reportC
Create a Touchstone report from imported or solved Touchstone curves.
Inputs: name (required), curves (required), solution, variations, +2 more. Returns: object
aedt_delete_itemC
Delete selected AEDT items through controlled PyAEDT delete methods.
Inputs: method (required), args, kwargs. Returns: object
aedt_design_summaryB
Return a compact summary of session, projects, setups, boundaries, and modeler objects.
Inputs: no named parameters. Returns: object
aedt_edit_design_notesC
Replace the active design notes text when the active app supports edit_notes.
Inputs: text (required). Returns: object
aedt_environmentA
Inspect Python and PyAEDT availability without launching AEDT.
Inputs: no named parameters. Returns: object
aedt_export_app_dataC
Export common AEDT design artifacts through PyAEDT export methods.
Inputs: export_kind (required), args, kwargs. Returns: object
aedt_export_diagnosticsC
Export convergence, mesh statistics, or profile diagnostics for a setup.
Inputs: export_kind (required), setup (required), variations, output_file, +1 more. Returns: object
aedt_export_field_plotC
Export a named AEDT field plot.
Inputs: plot_name (required), output_dir (required), file_name, file_format. Returns: object
aedt_export_icepak_summaryB
Export an Icepak summary report for objects, faces, or monitor-style quantities.
Inputs: output_dir, solution_name, summary_type, geometry_type, +4 more. Returns: object
aedt_export_matrix_dataC
Export Maxwell/Q3D matrices or Q3D equivalent circuits.
Inputs: export_kind (required), args, kwargs. Returns: object
aedt_export_reportC
Export an AEDT report to a file.
Inputs: report_name (required), output_path (required), file_format. Returns: object
aedt_export_touchstone_dataC
Export Touchstone data from HFSS or Circuit apps through PyAEDT.
Inputs: setup, sweep, output_file, variations, +4 more. Returns: object
aedt_get_antenna_dataC
Return HFSS antenna data through PyAEDT when the active app supports it.
Inputs: frequencies, setup, sphere, variations, +4 more. Returns: object
aedt_get_evaluated_valueC
Evaluate a variable or expression in the active AEDT design.
Inputs: name (required), units. Returns: object
aedt_get_fans_operating_pointC
Return Icepak fan operating-point data when supported by the active app.
Inputs: export_file, setup_name, time_step, design_variation. Returns: object
aedt_get_monitor_dataB
Return monitor data from solvers that expose get_monitor_data.
Inputs: no named parameters. Returns: object
aedt_get_nominal_variationC
Return the nominal variation string or dictionary from the active design.
Inputs: with_values. Returns: object
aedt_get_output_variableC
Read a PyAEDT output variable value.
Inputs: variable (required), solution. Returns: object
aedt_get_profileC
Return solve profile data for a setup or the active profile context.
Inputs: name. Returns: object
aedt_get_rcs_dataC
Return HFSS radar cross-section data through PyAEDT when supported.
Inputs: frequencies, setup, expression, variations, +3 more. Returns: object
aedt_get_setup_propertiesC
Return the PyAEDT setup object summary and its property dictionary.
Inputs: name (required). Returns: object
aedt_get_solution_dataC
Get solution data from the active AEDT design.
Inputs: expressions (required), setup_name, variations, kwargs. Returns: object
aedt_get_touchstone_dataC
Return Touchstone data from the active AEDT design when supported by PyAEDT.
Inputs: setup, sweep, variations. Returns: object
aedt_get_traces_for_plotC
Return available report traces from PyAEDT get_traces_for_plot.
Inputs: kwargs. Returns: object
aedt_get_variablesA
List project and design variables from PyAEDT VariableManager.
Inputs: no named parameters. Returns: object
aedt_hfss_operationC
Run an allowlisted HFSS boundary, excitation, RLC, scattering, or source operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_icepak_operationC
Run an allowlisted Icepak thermal source, opening, wall, monitor, fan, or mesh operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_import_cadD
Import CAD/layout files through supported PyAEDT import methods.
Inputs: input_file (required), import_kind, kwargs. Returns: object
aedt_import_datasetC
Import a 1D or 3D dataset file into AEDT.
Inputs: input_file (required), dimensions, name, is_project_dataset, +2 more. Returns: object
aedt_import_touchstone_solutionC
Import a Touchstone solution into a Circuit-style design when supported.
Inputs: input_file (required), solution. Returns: object
aedt_insert_designB
Insert a design into an AEDT project through the native Project API.
Inputs: design_type (required), design_name (required), solution_type, project_name. Returns: object
aedt_insert_far_fieldC
Insert an HFSS infinite-sphere far-field setup.
Inputs: kwargs. Returns: object
aedt_insert_near_fieldC
Insert a near-field definition through a controlled PyAEDT helper.
Inputs: field_kind (required), args, kwargs. Returns: object
aedt_list_apiC
List public methods and attributes on a PyAEDT or native AEDT object.
Inputs: target, module_name, include_private. Returns: object
aedt_list_projectsB
List AEDT projects and designs through the Desktop context.
Inputs: no named parameters. Returns: object
aedt_list_variationsB
List solved or available variation strings for a setup and sweep.
Inputs: setup, sweep. Returns: object
aedt_material_object_summaryC
Return conductor/dielectric names and object material properties.
Inputs: assignment, prop_names. Returns: object
aedt_materials_operationC
Run an allowlisted materials library operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_materials_summaryA
Return project material keys, material classes, and used materials.
Inputs: no named parameters. Returns: object
aedt_maxwell_operationC
Run an allowlisted Maxwell coil, winding, motion, force, torque, or source operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_mesh_operationD
Call a public mesh operation on app.mesh.
Inputs: method (required), args, kwargs. Returns: object
aedt_mesh_summaryA
Return mesh operation names, mesh operations, and initial mesh settings.
Inputs: no named parameters. Returns: object
aedt_modeler_operationC
Run an allowlisted modeler operation for transforms, booleans, sweeps, groups, or regions.
Inputs: method (required), args, kwargs. Returns: object
aedt_modeler_summaryA
Return modeler object names, units, coordinate systems, and bounding-box metadata.
Inputs: no named parameters. Returns: object
aedt_native_change_propertyC
Apply an AEDT native ChangeProperty payload to a target object.
Inputs: target (required), change_payload (required), module_name. Returns: object
aedt_native_get_propertiesC
Get AEDT native property names from a target object.
Inputs: target (required), tab (required), server (required), module_name. Returns: object
aedt_native_get_property_valueC
Get an AEDT native property value from a target object.
Inputs: target (required), tab (required), server (required), property_name (required), +1 more. Returns: object
aedt_native_module_batch_callC
Run ordered calls against one AEDT native module returned by odesign.GetModule.
Inputs: module_name (required), calls (required), continue_on_error. Returns: object
aedt_native_module_callC
Call a method on an AEDT native module returned by odesign.GetModule(module_name).
Inputs: module_name (required), method (required), args, kwargs. Returns: object
aedt_native_module_summaryC
Return callable method names and selected no-argument summaries for an AEDT native module.
Inputs: module_name (required), method_names, include_methods. Returns: object
aedt_new_projectB
Create a new AEDT project through the native Desktop API.
Inputs: project_name. Returns: object
aedt_oo_get_propertiesC
Return PyAEDT object-oriented property names for an AEDT object.
Inputs: target (required), object_name (required), module_name. Returns: object
aedt_oo_get_property_valueC
Return a PyAEDT object-oriented property value.
Inputs: target (required), object_name (required), property_name (required), module_name. Returns: object
aedt_oo_object_namesC
Return PyAEDT object-oriented child names for an AEDT object.
Inputs: target, object_name, module_name. Returns: object
aedt_oo_set_property_valueC
Set a PyAEDT object-oriented property value.
Inputs: target (required), object_name (required), property_name (required), value (required), +1 more. Returns: object
aedt_open_projectC
Open an AEDT project file in the active desktop session.
Inputs: project_path (required), design_name. Returns: object
aedt_optimetrics_setup_operationC
Run an allowlisted operation on an existing Optimetrics setup.
Inputs: collection (required), setup_name (required), method (required), args, +1 more. Returns: object
aedt_optimetrics_summaryA
Return parametric and optimization setup names from Optimetrics managers.
Inputs: no named parameters. Returns: object
aedt_optimization_operationD
Run an allowlisted Optimizations manager operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_parametric_operationC
Run an allowlisted Parametrics manager operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_post_operationC
Run an allowlisted post-processing operation for reports, fields, plots, or exports.
Inputs: method (required), args, kwargs. Returns: object
aedt_post_summaryA
Return report names, report types, field plots, and available post quantities.
Inputs: no named parameters. Returns: object
aedt_project_design_operationC
Run an allowlisted project/design maintenance operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_q3d_net_summaryC
Return Q3D net names, sources, sinks, and objects when available.
Inputs: net_name. Returns: object
aedt_q3d_operationC
Run an allowlisted Q3D/Q2D net, source, sink, thin conductor, or matrix operation.
Inputs: method (required), args, kwargs. Returns: object
aedt_read_design_dataB
Read PyAEDT design data dictionary for the active design.
Inputs: no named parameters. Returns: object
aedt_release_sessionC
Release the active AEDT session.
Inputs: close_projects, close_desktop. Returns: object
aedt_run_app_methodC
Call a public method on the active PyAEDT app object.
Inputs: method (required), args, kwargs. Returns: object
aedt_save_projectC
Save the active AEDT project.
Inputs: project_name, project_path, overwrite. Returns: object
aedt_session_infoA
Return active AEDT app, project, design, and version context.
Inputs: no named parameters. Returns: object
aedt_set_active_designC
Set the active AEDT design by name.
Inputs: design_name (required), design_type. Returns: object
aedt_set_active_projectC
Set the active AEDT project by name.
Inputs: project_name (required). Returns: object
aedt_set_hpc_optionsC
Apply custom HPC options or load an HPC configuration from an ACF file.
Inputs: cores, gpus, tasks, num_variations_to_distribute, +4 more. Returns: object
aedt_set_license_typeC
Set the active PyAEDT license type when supported by the active app.
Inputs: license_type. Returns: object
aedt_set_temporary_directoryC
Set the active PyAEDT temporary directory for generated files.
Inputs: path (required). Returns: object
aedt_setup_summaryA
Return setup names, active setup, nominal sweep, and setup sweep names.
Inputs: no named parameters. Returns: object
aedt_set_variableC
Set a design variable or project variable. Prefix name with '$' for project variables.
Inputs: name (required), expression (required), description, sweep, +2 more. Returns: object
aedt_signal_integrity_expressionsC
Return return-loss, insertion-loss, NEXT, and FEXT expression lists when supported.
Inputs: drivers, receivers, excitations, drivers_prefix_name, +4 more. Returns: object
aedt_solve_in_batchC
Launch PyAEDT batch solve for the active design or a named setup.
Inputs: file_name, machine, run_in_thread, cores, +3 more. Returns: object
aedt_source_port_summaryA
Summarize available ports, sources, and source modes from the active PyAEDT app.
Inputs: no named parameters. Returns: object
aedt_start_sessionC
Launch or connect to an AEDT application through PyAEDT.
Inputs: app_name, version, project, design, +9 more. Returns: object
aedt_update_configuration_optionsC
Set configuration export/import options or run an all-options action.
Inputs: options, action. Returns: object
aedt_update_setupC
Update an existing AEDT setup property dictionary and call setup.update().
Inputs: name (required), properties (required). Returns: object
aedt_validate_designC
Run PyAEDT simple or full design validation.
Inputs: validation_kind, args, kwargs. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 109 tools
Many tools have overlapping purposes, especially the generic escape hatches: aedt_run_app_method, aedt_call, aedt_batch_call, aedt_native_module_call, and aedt_native_module_batch_call all expose similar 'call any method' behavior in slightly different contexts. Specific operations also blur boundaries, e.g. aedt_hfss_operation vs aedt_assign_boundary_or_excitation vs aedt_create_port, making it hard for an agent to confidently select the right tool.
Tool names consistently follow an aedt_ prefix with snake_case and a verb_noun pattern (create_, get_, set_, list_, export_, import_, update_, analyze_). Even compound nouns like aedt_source_port_summary and aedt_optimetrics_setup_operation remain predictable, and there is no mixing of conventions.
109 tools is an extreme count for any MCP server, far exceeding the 25+ threshold for poor scoping. While the AEDT domain is broad, this many tools creates a massive selection space that will overwhelm agents and defeat the purpose of a curated tool surface.
The tool surface is unusually broad, covering projects, designs, setups, sweeps, boundaries, materials, meshing, post-processing, exports, and native API access. The generic fallback tools (aedt_call, aedt_batch_call) ensure no operation is a hard dead end, though some specialized workflows still depend on those generic calls rather than purpose-built tools.
Maintenance
Related MCP Connectors
Persistent memory, knowledge graph, task management and more for AI agents via MCP.
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Protocol-native energy infrastructure orchestration for AI data centers. Provides 46 MCP tools across 8 grid protocols (IEC-61850, DNP3, Modbus, OCPP, OpenADR, IEEE 2030.5, IEC 60870-5-104, ICCP) with 5 core API primitives: connect, dispatch, settle, comply, and intel. Enables AI agents to programmatically interact with substations, grid interfaces, and energy assets for real-time workload-grid coordination.
LLM Orchestration Agent (Mcp)
Related MCP Servers
- AlicenseBqualityBmaintenanceEnables AI agents to automate multiphysics simulations in COMSOL Multiphysics, covering model management, geometry building, physics configuration, and results visualization. It supports complex simulation workflows through the MCP protocol and includes integrated knowledge retrieval for documentation and troubleshooting.93959MIT
- AlicenseBqualityCmaintenanceEnables AI agents to automate COMSOL Multiphysics simulations, including model management, geometry building, physics configuration, meshing, solving, and results visualization through the MCP protocol.78MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to interact with Ansys simulation software (Fluent, MAPDL, Mechanical, Geometry) through the Model Context Protocol.71MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to automate COMSOL Multiphysics simulations, including model management, geometry building, physics configuration, meshing, solving, and results visualization via the MCP protocol.MIT