AdminConsole
Platforms/MCP
AI assistants, copilots, and spatial workflow agents

MCP

Give AI systems typed Mappls capabilities with policy, evidence, and human control.

TypeScriptPythonJSON Schema
Open quickstart Browse source guides
Recommended boundary

How the pieces fit

Typed tools and resources that let agents discover and invoke Mappls capabilities safely. Use this boundary as a starting point, then narrow it around the exact Mappls products and trust model your application needs.

01

AI host

Plan user intent, request tool calls, present approvals, and keep generated text separate from authoritative results.

02

Mappls MCP boundary

Advertise narrow schemas, validate arguments, enforce policy, call supported Mappls services, and return provenance.

03

Credential and audit plane

Scope credentials by tenant and tool, redact logs, meter usage, trace calls, and support revocation.

A safe first integration

From empty project to one real journey

Each step produces a testable boundary. Keep the first release deliberately small, then expand from observed product behavior.

  1. 1

    Start read-only

    Expose discovery, search, route, and documentation tools before any command that changes durable state.

  2. 2

    Constrain every schema

    Use enums, bounds, required fields, coordinate validation, result limits, and explicit units.

  3. 3

    Separate plan from effect

    Preview stateful commands, require approval, attach idempotency, and return the created resource identifier.

  4. 4

    Evaluate adversarially

    Test prompt injection, secret requests, confused-deputy access, fabricated APIs, retries, and partial failure.

Capabilities on MCP

2 product families

2 of these families include durable or session state. Follow their complete lifecycle guidance rather than integrating only the first successful call.

hybrid

GIS & Analytics

Spatial data, analysis, maps, dashboards, and embeddable location intelligence.

6 capabilities
hybrid

AI & Location

AI/ML APIs, vision, location-aware agents, ranking, and natural-language spatial workflows.

6 capabilities
Auditable implementation evidence

SDKs, APIs, distributions, and samples

Every item below is generated from a public Mappls repository and links to its source-grounded guide. Maturity labels distinguish current material from samples, distributions, legacy sources, and deprecated integrations.

REST API

MAPPLS AI / ML APIs

mappls-ai-apis

Revolutionize spatial analysis with our AI-ML API's, delivering unparalleled insights and predictive capabilities for optimized decision-making.

REST API

Skydnn AI APIs

skydnn-aiapi-docs

A Mappls rest api for maps, gis analytics, ai vision, capture feedback across REST, AI/MCP.

Learn by building

Platform tutorials

Start with one focused result, then move into complete stateful journeys and operational failure handling.

Intermediate · 30 min

Give an AI agent grounded Mappls tools

An agent that can reason about real places without inventing APIs.

Open tutorial
Advanced · 55 min

Provision and validate a remote MCP client

A replay-safe client identity journey with durable registration, request checks, revocation, and audit evidence.

Open tutorial
Before launch

Production readiness

Pass these checks in addition to the requirements of the selected product, SDK release, operating system, and security program.

  • Tools expose the least authority needed
  • State-changing calls require explicit policy and approval
  • Responses identify source, freshness, and uncertainty
  • Every tool call is tenant-scoped, metered, redacted, and traceable