- Home
- MCP servers
- QuantumArchitect
QuantumArchitect
- python
0
GitHub Stars
python
Language
6 months ago
First Indexed
3 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
QuantumArchitect MCP Server provides a programmable quantum circuit engine that AI agents can call to create, validate, simulate, and score quantum circuits. It exposes MCP-compliant endpoints so you can embed quantum reasoning into your agents and workflows with practical templates, templates, and hardware-aware validation.
How to use
You connect your MCP-enabled AI agent to the QuantumArchitect MCP HTTP endpoint to access creation, validation, evaluation, and scoring tools. Start by starting the server locally or hosting it in your environment, then point your agent to the provided MCP URL. Use the creation tools to generate circuits, apply validation to ensure hardware compatibility, run simulations to inspect statevectors and noise, and finally obtain a circuit score to guide decisions.
How to install
Prerequisites: Python is installed on your system. You will also need pip to install dependencies.
python --version
pip --version
- Create a project directory and navigate into it.
mkdir quantumarchitect_mcp
cd quantumarchitect_mcp
- Install the required Python packages from the project’s dependencies file.
pip install -r requirements.txt
- Start the MCP server application.
python app.py
- Open the web interface to verify the server is running and accessible at the default address.
# The UI runs by default at http://127.0.0.1:7861
Additional content
Configuration and MCP integration details are provided below to help you connect your AI agents and customize the server behavior. You can expose the server via an HTTP MCP endpoint so clients can invoke creation, validation, evaluation, and scoring functions.
MCP endpoints and integration notes
The server exposes a set of MCP endpoints that your agent can call to perform circuit-related tasks. You can connect an MCP client by configuring the server URL in the client’s MCP settings.
Available tools
mcp_create_circuit
Create a circuit from a predefined template such as bell_state, ghz_state, qft, grover, or vqe ansatz.
mcp_parse_qasm
Parse OpenQASM code to validate syntax and extract circuit structure.
mcp_build_circuit
Assemble a circuit from a description of gates and measurements.
mcp_validate_circuit
Validate a circuit against syntax, hardware target, and unitary properties.
mcp_check_hardware
Check compatibility of a circuit with a specific hardware topology.
mcp_simulate
Run statevector simulations and return amplitudes, probabilities, and optionally the full statevector.
mcp_get_statevector
Return the ideal statevector for a given circuit.
mcp_estimate_fidelity
Estimate fidelity against a target hardware profile.
mcp_score_circuit
Provide a score for a circuit based on complexity, expressibility, and hardware fitness.
mcp_compare_circuits
Compare multiple circuits on a given hardware profile.
mcp_get_gate_info
Return documentation and details for a specific quantum gate.
mcp_get_algorithm_info
Explain a quantum algorithm implemented by a circuit.
mcp_list_hardware
List available hardware profiles supported by the server.
mcp_list_templates
List available circuit templates for quick generation.
mcp_get_learning_path
Provide learning resources aligned to the current proficiency level.