Example Session
tool: collections.list
tool: schema.describe
input:
collection: utilities.water
tool: features.query
input:
collection: utilities.water
where: status = 'active'
within: district_7AI Agents
Let AI workflows work with spatial context directly.
Large language models are good at reasoning, but they still need a clear interface for layers, schemas, geometry filters, and spatial rules. Honua exposes that interface through the Model Context Protocol (MCP) — an open standard for connecting AI models to external tools and data sources — so automation can operate on maps and datasets without ad hoc wrappers.
Discoverable collections
Schema inspection
Filtered queries
Topology-aware tools
Why It Matters
Without a structured tool layer, AI integrations end up guessing at endpoints and spatial fields. MCP gives the model a typed contract it can discover at runtime.
What Teams Get
Build AI workflows on real geospatial structure instead of hand-written wrappers.
Discovery
Show agents what data exists before they try to use it.
Collections, layers, and service capabilities can be discovered directly so agents do not need hidden assumptions about what your runtime contains.
Understanding
Expose fields, geometry types, and constraints in a structured format.
Schema inspection helps automation understand the difference between parcels, utilities, addresses, and imagery-backed layers before it generates a query.
Action
Run queries and spatial checks without sending humans through a manual chain.
Use filtered retrieval, spatial predicates, and topology checks as first-class tools inside automated workflows.
Why Not Just Use REST?
General APIs are useful. They are not enough on their own for governed AI work.
REST endpoints can move data, but AI systems still need a predictable way to discover what they can do, what each dataset means, and which operations are safe to call.
Common Failure
Agents guess at endpoints, fields, and geometry behavior.
That leads to fragile prompts, wasted calls, and silent mistakes when datasets change shape or when layers do not behave the same way between environments.
Honua Approach
Give the workflow a tool contract that explains itself.
The MCP layer lets agents discover available tools, inspect the shape of the data, and perform spatial work through a governed interface instead of a pile of assumptions.
Typical Uses
Where teams start using AI agents with spatial data.
Support and Search
Answer operational questions with map-backed context.
Let assistants inspect service health, list available layers, and retrieve map or feature context when a user asks about a place or asset.
Planning
Check spatial rules before work moves forward.
Use buffer checks, overlap checks, and collection queries to support planning, review, and permit-related workflows.
Automation
Connect LLM workflows to governed spatial actions.
Route tasks from internal copilots or agent frameworks into spatial tools without exposing an unmanaged API surface to every workflow author.
The difference between an AI demo and an AI workflow is whether the model can discover
what tools exist, understand what each one expects, and get a predictable response back.
Next Step
See how the runtime and SDKs support AI workflows.
The gRPC runtime handles the transport and the SDKs handle the integration. Start with the docs if you want to get hands-on.