Limen is designed as open-source civic and scientific infrastructure for landslide-risk monitoring. It combines public geospatial and environmental datasets with interpretable scoring, PostGIS, APIs, maps and notification workflows.
Landslide-risk information is often distributed across heterogeneous datasets, portals and formats. Limen aims to make that information easier to operationalize by combining:
- a 1 km² grid model over Italian territory;
- public hazard, morphology, rainfall, soil-moisture, seismic and fire-history signals;
- interpretable multi-factor scoring;
- PostGIS-native storage and vector-tile serving;
- API and frontend map access;
- MCP tools for agentic geospatial workflows;
- deterministic reports and alerts that show why risk changed.
- Researchers who need reproducible geospatial risk workflows.
- Civic-tech groups working with environmental and territorial data.
- Public administrators and technical offices evaluating monitoring prototypes.
- Developers building PostGIS, map and MCP-based geospatial tools.
- Journalists and local communities exploring environmental-risk information.
Limen is not only an application. Its most reusable OSS parts are:
- PostGIS grid and AOI modelling patterns;
- geospatial ingestion adapters;
- interpretable scoring configuration and thresholds;
- risk explanation and cell-breakdown APIs;
- vector-tile and MapLibre frontend patterns;
- notification and alert-deduplication workflows;
- MCP tools for risk summaries and operational queries.
Limen is a monitoring and decision-support framework, not an official emergency-warning system unless validated and adopted by an authorized institution. Public communication should keep uncertainty visible and separate official-source data from derived/modelled indicators.
Claude can help maintain the project by improving documentation, writing tests, reviewing scoring changes, checking runbooks and making the codebase easier for external contributors. AI-generated briefing text should remain grounded in deterministic model outputs and public data sources.
The next OSS goal is to make Limen easier to evaluate and reuse by external contributors. Useful indicators include:
- external issues and pull requests;
- reproducible setup by someone outside the maintainer team;
- documented reuse of geospatial modules or MCP tools;
- public validation reports and demo datasets;
- community contributions to datasets, docs, scoring tests or frontend map UX.
Limen should be positioned as an early-stage but serious open-source geospatial monitoring framework. It may not yet meet numerical thresholds such as many dependents or external contributors, but it has clear civic/scientific value, public-data grounding, interpretable scoring and reusable infrastructure for geospatial AI workflows.