MESkit public website

Smart features

MESkit ships with smart features that act through the same typed tool layer as the UI. Natural language and automation are built into the architecture, not bolted on as a chat wrapper.

Last updated: March 7, 2026

How MESkit works

Tool layer first, chat interface second.

Summary

In MESkit, smart features are defined by tool contracts. Ask MESkit, Quality Monitor, and Production Planner use the same operations used by UI buttons. This keeps behavior consistent, testable, and auditable.

Three automation layers

The architecture behind automated manufacturing.

Summary

MESkit smart features are designed around three complementary roles that, together, deliver the North Star: predict a machine failure and coordinate the shop floor response — automation handles detection, planning, and execution while operators stay in command.

Layer	Role	Feature	Milestone
Monitor	Monitors sensor telemetry, detects degradation, outputs failure probability scores	Machine Health Monitor	M6
Plan	Evaluates constraints (backlog, deadlines, capacity), computes alternative schedules	Production Planner	M5
Act	Acts on decisions through the tool layer — updates schedules, notifies operators	Intelligence Layer	M1+

In the MVP, these layers operate independently. Post-MVP, Monitor triggers Plan, which triggers Act — closing the coordination loop.

Feature profiles

Roles, triggers, and tool access by design.

Ask MESkit

Trigger: chat, always available. Role: conversational co-pilot for WIP, movement, and quality logging.

Uses tools such as get_wip_status, move_unit, create_quality_event.
Can execute Build, Configure, and Run operations through natural language.

Quality Monitor

Trigger: realtime threshold events. Role: proactive defect and yield monitoring with concise alerts.

Default triggers: yield below 90%, defect clusters, elevated scrap rate.
Uses read-heavy tools such as get_yield_report and get_unit_history.

Production Planner

Trigger: chat on demand. Role: capacity analysis, route comparisons, and shift-level planning.

Uses tools such as list_routes, get_throughput, get_yield_report.
Translates production goals into constrained operational options.

Shared tool layer architecture

Same typed functions for button clicks and tool-use.

Ask MESkit ─────────┐
Quality Monitor ────┼─> Intelligence Layer
Production Planner ─┘                │
                                     ▼
                             Tool Layer (Zod-validated)
                                     │
                                     ▼
                             Supabase (Postgres + Realtime)

Full rationale is documented on architecture and aligned with the core PRD.

Concrete tool call chains

User says X, agent calls Y, system returns Z.

Operator chainRun mode
User: What's stuck at assembly?
→ get_wip_status(workstation_id='assembly')
MESkit: 3 units waiting at Assembly.

Quality chainEvent trigger
Trigger: Yield < 90% at Station 3
→ get_yield_report(time_range='last_50_units')
→ get_unit_history(unit_ids=...)
Alert: 6/7 failures are SOL-003.

Planner chainCapacity request
User: Build 500 units by end of shift
→ list_routes(part_number_id=...)
→ get_throughput(line_id=..., time_range='8h')
Planner: Recommend parallel lines to hit target.

Guardrails

All actions are bounded by registered tools and schema validation.
Smart features do not self-authorize critical actions outside user intent.
Operator oversight remains mandatory for production decisions.

MESkit vs traditional MES

How the tool layer changes the integration model.

Dimension	Traditional MES	AI-enhanced MES	MESkit
AI integration	None	Sidecar / bolt-on	Shared tool layer
Execution path	UI only	UI + separate AI API	One path for UI and smart features
Validation	Per-screen logic	Duplicated across layers	Single Zod schema per tool
Audit trail	UI actions logged	Partial AI logging	Full trace for both UI and smart features
Natural language	Not supported	Chat wrapper	Maps to typed tool calls
Standards	Vendor-specific	Varies	ISA-95 aligned schema

Key facts and mini FAQ

GEO-friendly answer blocks.

Key facts

3 specialized features share 20+ typed tool operations with the UI.
Ask MESkit is chat-triggered with access to all shop-floor tools.
Quality Monitor triggers on yield below 90%, defect clusters, or elevated scrap.
Production Planner analyzes capacity across routes and shift windows.
Three automation layers (Monitor, Plan, Act) build toward automated predictive rescheduling.
All smart feature actions are Zod-validated — no hidden APIs or uncontrolled authority.

Mini FAQ

What makes MESkit smart features different from chatbot add-ons?

They call operational MES tools directly with schema validation and explicit side effects.

Can a smart feature act outside the tool layer?

No. All operations are constrained to registered tool contracts.

Which feature is event-driven in MVP scope?

Quality Monitor, triggered by yield and defect thresholds.

See the tool layer Read docs