Band 1
Free Conversation
Greetings, smalltalk, generic questions get a natural answer. Off-topic is welcomed; MH AI politely notes that its primary expertise is Macro Handler.
AI Assistant
MH AI is a smart assistant that can chat naturally, knows the app, and designs macros for you through an approval gate. Pick from cloud providers such as Gemini, OpenAI, Claude, Groq, DeepSeek, Qwen, and more as the brain (BYO API key); on-device local models are on the roadmap.
Depending on your message, MH AI picks a different prompt family. General chat is light; an app question is grounded; an automation request is structured and approval-gated.
Band 1
Greetings, smalltalk, generic questions get a natural answer. Off-topic is welcomed; MH AI politely notes that its primary expertise is Macro Handler.
Band 2
Questions like "how to use Region" or "OCR in no-code" are answered from ScriptApiCatalog and VisualBlockRegistry. The model cannot invent APIs.
Band 3
"Build me a macro" requests produce a structured JSON proposal. Validator → risk → preview → YOUR approval → apply → rollback if needed.
In the current release path, you can choose from 14 cloud providers with your own API key. The on-device local model catalog is on the roadmap; once release-ready, it will use the same capability registry, safety rules, and approval flow.
Cloud — high capacity, multimodal. API key is BYO (bring your own).
Cloud — structured macro proposal via function-calling. BYO key.
Cloud — stable across long contexts, safety-aware. BYO key.
Cloud — low-latency high-speed inference. BYO key.
Cloud — code-oriented models. BYO key.
Cloud — Alibaba Qwen models (OpenAI-compatible endpoint). BYO key.
Cloud — xAI Grok models. BYO key.
Cloud — multi-provider routing behind one key. BYO key.
Cloud — high-speed inference. BYO key.
Cloud — open-weight model hosting. BYO key.
Cloud — search-grounded answers. BYO key.
Cloud — Mistral models. BYO key.
Cloud — Together AI model hosting. BYO key.
Cloud — Cohere Command models. BYO key.
On-device local models (coming soon). Will run without internet once released.
MH AI auto-classifies your message; each class routes to the right pipeline.
Greetings, smalltalk, off-topic — natural answer; Macro Handler primary expertise is politely noted.
Questions about API, blocks, no-code/code editor. Grounded answer from ScriptApiCatalog + VisualBlockRegistry.
Natural-language macro brief → structured JSON proposal. Only registered capability IDs are used.
Lua script for the code editor or block chain for the no-code workspace. Everything flows through approval.
Inspects your active macro, surfaces issues, proposes fixes — you approve before anything applies.
AI learns while you act or assists live. Still routed through the approval gate.
AI cannot hallucinate, bypass your approval, or propose risky actions. Every rule is locked in by tests.
AI only produces a plan. CodeEditorWriter, NoCodeWorkspaceWriter, and AndroidControlLayer require approval.
If a proposal references an unknown capability id, the safety gate rejects it; after a single retry, a clarifying question is shown.
Denylisted actions (payment, account deletion, root, wipe) cannot be proposed by any provider.
If the provider claims it already applied or ran the macro, that approval-bypass false-claim is rejected.
When the model reports low confidence, MH AI asks what you actually want instead of auto-applying.
When AI produces a proposal, this is the path it takes — nothing moves without you.
Open the MH AI panel and type something like "find Y in app X, tap it, wait 1 second, repeat". AI proposes the right block chain; you approve, and it lands in the no-code workspace.
Go to no-code →Get a Lua draft from a sentence like "find a template in Region, tap if found, wait 500 ms". AI uses code from ScriptApiCatalog while the sandbox boundary stays intact.
Go to code editor →After installing, the MH AI panel is one tap away on every screen. Pick a provider, paste your key, start building macros.