AI Generated Model Context Protocol (MCP) server for geospatial (Ordnance Survey) tooling built with FastAPI & Python 3.11+. Provides a uniform tool abstraction, typed schemas, structured error model, correlation IDs, and high test coverage
This repository also configures its own agents. See what mcp-geo tells them →
Skill Claude CodeCodex
This document describes how to use the MCP Geo server for UK geospatial, ONS, and NOMIS statistics workflows. It is intended for MCP clients and agents as a quick, reliable guide to tool selection.
Skill Claude CodeCodex
Build or refine user-facing mcp-geo HTML maps with OS vector detail, label-safe overlays, browser-stored OS keys, and real-browser validation.
Skill Codex needs its repo
Generate a Long Horizon-style Codex metrics summary for mcp-geo from local Codex session logs.
Skill Claude CodeCodex needs its repo
Refresh, validate, and maintain the checked-in MCP Geo Obsidian knowledge base, including the canonical vault, local overlay, and drift checks.
Skill Claude CodeCodex needs its repo
Analyze mcp-geo repository extent and functional complexity using dual-scope inventory, generated-output exclusion, hotspot scoring (complexity x churn), and optional GitHub stats.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: