YonSci/climate-agent

climate-agent is a command-line Python tool that automates the full pipeline for acquiring, processing, and validating climate datasets — primarily for East Africa impact modelling at ILRI (the International Livestock Research Institute). # Climate Data Harmonization Agent ## Climate Data Harmonization Agent for Automated Acquisition, Standardization, Validation and Delivery of Climate Data **Institutional context:** Developed for ILRI Climate Services to support climate-risk analysis, impact modelling, and harmonized climate-data preparation for livestock, agriculture, food security, drought, disease risk, and related decision-support applications. **Project status:** Operational research software, version `0.1.0`. The repository demonstrates product development, documentation, reproducibility, validation, public sharing, and a live evidence dashboard. --- ## Live dashboard ### Open the Climate Data Harmonization Dashboard *Dashboard overview showing operational run statistics, supported coverage, climate variables, data sources, and quality-assurance information. Select the image to open the live dashboard.* ### Automated validation, diagnostics, and run reports View the data inventory *Representative inventory view showing country, variable, scenario, period, processing status, run identifier, date, and duration.* View the run history *Representative run-history view showing completed workflows, requested variables and scenarios, execution duration, and stage-level outcomes.* The GitHub Pages dashboard is the public, human-readable evidence interface for the tool. It is generated from committed run manifests and associated diagnostic products and can be used to review: - the current climate-data inventory; - completed, warning, and failed processing runs; - countries, variables, scenarios, periods, and run identifiers; - stage-level status, commands, retries, and execution metadata; - schema, unit, temporal, spatial, and anomaly checks; - quality-control statistics and diagnostic plots; and - evidence that outputs have been processed, validated, and documented. The screenshots above are representative captures and may show historical generation dates or run stati …