What it does
The server, published by below280 (described as the UK partner for openLCA), gives AI assistants access to a running openLCA instance through 35 tools. They cover database exploration, building and editing models, calculations such as scenarios, sensitivity analysis, Monte Carlo runs and contribution analysis, auditing and validating models, and extracting data quality assessments. It works with ecoinvent-family and FLCAC-family databases, either asking which family is in use or detecting it from flow property names.
It suits people working on life cycle assessment who want to query a database or build and calculate models through a conversation. Example requests include building a model from a set of inputs, comparing scenarios with different transport distances, finding the processes that contribute most to climate change, varying parameters, and running Monte Carlo iterations to view uncertainty. The AI assistant handles the conversation and presents results with charts, tables and exportable data.
The server runs locally over stdio transport and needs a local Python installation, set up either from PyPI or from a GitHub clone. It is configured in an MCP client such as Claude Desktop, Cursor or VS Code. To use it, open a database in openLCA, start the IPC server on port 8080, then restart or reconnect the client. After changes, refresh openLCA's GUI to see them.
Limitations and requirements
Requires a running openLCA instance with the IPC server started on port 8080, and a local Python installation. The server is beta software: test it on a copy of the database before connecting a production database. The AI client sends tool results, including process names, exchange data, parameter values and impact results, to its provider, so do not connect a confidential database through a personal or free-tier AI account; check the provider's data processing terms. ChatGPT support is experimental and requires a Secure MCP Tunnel. Claude Desktop on Windows may need the full path to the Python executable.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
openLCA MCP Server
Connect AI assistants to openLCA. 35 tools for database search, model building, and LCIA.
v1.16.0As shown on registry.modelcontextprotocol.io when we last read it