Forge Suite · v1.2.5

ForgeMind™

The GIS Intent Compiler

Natural language in. Correct geospatial pipelines out. Or an honest refusal. ForgeMind is the natural-language control plane for ForgeGIS — it turns a plain-language request into a validated, executed pipeline on the GPU-accelerated engine, and when a request cannot be honored correctly it returns an explicit refusal rather than a confident guess.

Zero
confident-wrong results across thousands of turns in the June 2026 validation campaign
~25×
less model spend than a legacy agent-loop run on one site-suitability request (Claude, May 2026)
Your model
Claude, OpenAI, Gemini, Grok — or a model server you run yourself
MCP
native in both directions: it drives tool servers, and it is one

What it is, why it matters, how it's built differently

Three short answers for the technical evaluator.

What it is

ForgeMind sits between a human (or another agent) and the ForgeGIS™ compute engine. It decides what to run, binds the request to real data, dispatches the work, and narrates the result — turning informal language into a correct, reproducible operation. It speaks the Model Context Protocol natively in both directions.

Why it matters

A conventional agent lets a model author tool chains directly, so every wrong parameter and hallucinated step lands in production looking confident. On its compiled path, ForgeMind refuses by construction: a request either compiles to a validated pipeline or is declined. That is what regulated and mission work requires.

How it's built differently

The language model classifies intent and extracts parameters only. Deterministic code owns pipeline structure, type-checking, and execution — so the same request yields the same pipeline every time, with no hallucinated params and no drift. Cheap work runs on a fast model tier; the costly tier is reserved for the hard step.

The compiler model

ForgeMind treats a natural-language request the way a compiler treats source code. The language model does one narrow job — extract intent into a typed intermediate representation. From there, deterministic machinery takes over.

Natural language
“Compute least-cost path from ingress_ip to hlz_charlie”
LLM: intent extract
Maps words to a typed intent. No execution.
Typed IR
Operation + resolved anchors + parameters
Deterministic check + route
Type-check, resolve refs, refuse if unsound
GPU execution
ForgeGIS runs the validated operation
Typical LLM agent loop ForgeMind compiler
What the LLM does Plans and executes, reasons about geometry directly Extracts intent into a typed IR, nothing more
Where math happens Inside the model's reasoning (unverifiable) Deterministic engine · GPU-accelerated ForgeGIS
Failure mode Plausible, confident, wrong Explicit refusal. Never a fabricated number
Cost / latency Many model calls per task One intent pass, then deterministic execution
Auditability Opaque chain of thought Inspectable IR for every single turn

Architecture commitments

The internal GIS Intent Compiler is proprietary; what follows is the externally observable contract each commitment produces.

Refusal by construction

Validation before execution

Every candidate pipeline is checked against a type system and operational guards before any work is dispatched. A request that cannot be mapped to a valid operation, lacks its parameters, or cannot bind to data is refused with a specific, machine-readable reason — never run speculatively. A request the compiler cannot route may instead fall back to a guarded agent loop, where the model chooses the tool calls; structural failures and out-of-scope requests are refused outright.

Narrow the model

The LLM does only intent

The language model is confined to classifying what the operator wants and extracting parameters. The shape of the pipeline is owned by deterministic machinery, so for a templated capability the same parsed intent produces the same pipeline every time — no hallucinated parameters in the pipeline, no operation-arity slips.

Cost engineering

The right model for each job

With a hosted vendor, cheap, high-volume classification runs on a fast, low-cost tier and the more capable tier is kept for requests no template covers; most answers are now narrated from templates with no model call at all. On one site-suitability request (Claude, May 2026) the compiled path cost roughly a twenty-fifth of a legacy agent-loop run, which hit its iteration cap without finishing. In the suite a local model runs every stage on one model, on your own hardware.

No vendor lock-in

LLM-agnostic by design

The provider and model sit behind one abstraction — configuration, not architecture. Switch models to follow cost, capability, availability, or procurement constraints (including continuity during a provider outage) without code changes. Claude, OpenAI, Gemini, and Grok ship built in, and a model you host yourself on any OpenAI-compatible server — Ollama is the tested path — plugs into the same integration point.

MCP citizen

Native MCP, both directions

As a client it dispatches to MCP tool servers — the ForgeGIS compute surface, a dataset catalog, a map client — guarding each namespace behind a circuit breaker. As a server it exposes a single high-level tool a higher-order agent can hand an entire geospatial workflow to, plus a structured surface — list_capabilities and run_capability — that runs capabilities with no model call at all.

New in 1.2.5

It answers on the map you asked from

Ask in ForgeGIS Studio and the contours, profiles, and draped rasters land on Studio's map; ask in Seaglass Globe's Assistant and they land on the globe. Each request carries the window it came from, along with the scene you were looking at, so the answer goes back to it.

Geometry the agent can reason about

Named scene entities

Named geometry a user places on a map — a dock, a route, an objective — is referenced by a stable handle and hydrated from the catalog on demand, never re-derived from a bounding box. Fail-closed: if the true geometry is not reachable, the operation defers and asks rather than emitting a plausible-but-wrong product.

Showcase · Session S-MARATHON-22

One conversation. A complete mission plan.

A scripted test scenario simulating a combat search-and-rescue plan, run end to end against real SRTM terrain. The test harness supplies each turn's intent, so what it exercises is routing, type-checking, entity memory and execution — and all 102 operations succeeded.

102
operations, one session
0
failed operations
Read the full marathon showcase →

Capabilities, by intent

ForgeMind makes the ForgeGIS catalog reachable in natural language, anchored on a deterministic core of routed capabilities. Coverage grows capability by capability as each is validated; ForgeMind advertises only what it can run correctly.

Direct analysis. “Compute viewshed from 34.5N 69.2E with a 10 km radius.” — explicit-coordinate raster analysis.
Discovery into analysis. “What elevation datasets do we have for this region? Pick the highest-resolution one and run a slope analysis.” — a catalog query chained into a compute operation.
Site suitability. “Where can we land outside the enemy viewshed?” — a multi-operation suitability workflow combining line-of-sight exclusion with terrain constraints.
Scene-aware. “Mark Objective Bravo here and map the slope around it,” then later “rebuild the slope picture — it displaced to a new grid reference” — named-anchor placement and refinement across turns.
Visualization. “Load the result onto the globe as a red, 50% transparent layer.” — chaining a compute product into a map-client display.

Validation and coverage

ForgeMind is validated along two complementary axes — a large multi-turn conversation campaign that measures end-to-end behavior as an operator would experience it, and a focused compiler-evaluation campaign that measures the core intent-to-pipeline transformation in isolation.

Full validation methodology and campaign detail: ForgeMind Technical Brief (PDF).

Deployment and integration

ForgeMind ships as a single self-contained Java application, deployable where the mission runs.

Two shapes, one artifact

Run it as an HTTP service exposing a small REST surface — a buffered endpoint plus a streaming (SSE) endpoint that surfaces each model round trip and tool call live — or as an MCP stdio server that presents ForgeMind as a single delegatable tool to a higher-level agent host. Auth, rate limiting, request-size caps, and a concurrency cap are built in.

Purpose-built for ForgeGIS

ForgeMind spawns and supervises its MCP backends — the ForgeGIS compute engine, a dataset catalog, and a map client — as local subprocesses over stdio, with health checks and per-namespace circuit breakers. ForgeGIS is the engine it is built to drive; the catalog and map-client roles are pluggable behind the protocol.

Fit for regulated & air-gapped sites

A pure-JVM application with local-disk persistence and no required external services beyond the language-provider endpoint. The memory store is a single local database file; backends run over stdio, not the network. An air-gapped site points ForgeMind at a model server on its own network — any OpenAI-compatible server, with Ollama the tested path — so no request goes to a cloud model at all.

Who it's for

One product, three readers. Each one-pager leads with the parts of ForgeMind that matter most to that audience.

Downloads

All ForgeMind collateral. Direct download — no email gate, no form wall.

Talk to us about ForgeMind

Licensing questions, evaluation access, partnership and distribution inquiries — we read every email. ForgeMind is commercially licensed; terms are provided on request, and an evaluation build is available to qualified teams.

rich@seaglassfoundry.com