mcpdirectory.dev
9,129 servers · updated 23 Sept

Perception MCP server

community Active last commit 18 d 
{
  "mcpServers": {
    "geiant-perception": {
      "type": "http",
      "url": "https://packagesmcp-perception-production.up.railway.app/mcp"
    }
  }
}
file .mcp.json (in your project root)version 0.3.3

What it does

GEIANT is a geospatial AI agent orchestration runtime where agents (called 'ants') are cryptographically identified and bound to H3 grid territories. Tasks are routed through a compliance enforcement layer that validates jurisdiction, delegation chains, geometry, and signatures before dispatch. It is for developers building multi-agent systems that need location-aware identity and regulatory compliance.

  • Route multi-agent tasks through jurisdiction-aware compliance gates that enforce GDPR, EU AI Act, and other regulatory frameworks before execution.
  • Build collaborative workflows where agent identity is bound to H3 geospatial territories and tracked through an immutable audit log (Virtual Breadcrumb chain).
  • Validate geospatial task inputs for self-intersections, coordinate errors, and unclosed geometries before passing to downstream agents.

Tools this server exposes

ToolDoes
registryQuery the Agent Registry (Nest) to list available ants and their capabilities.
jurisdictionResolve an H3 cell to its geographic location, country, and applicable regulatory frameworks.
routeSubmit a signed task for routing through the four compliance gates: signature verification, jurisdiction resolution, delegation chain validation, and geometry p
healthHealth check endpoint for the GEIANT API.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.