Built for careful adoption
Companies want to know what the AI may do, who is responsible, and what record is kept. Naikaku keeps the decision before and the record after.
Japan-built · Japanese-primary · Cautious AI adoption
Naikaku is a workbench for handing work to AI with separate roles: one plans, one does the work, others check it. Everything it did, and what came of it, can be reviewed afterwards.
No single AI gets the whole job. Planning, doing, critique, supervision, safety checks, memory and scoring are separate roles. Work only runs after the setup is checked, and people can review the logs and results afterwards.
Companies want to know what the AI may do, who is responsible, and what record is kept. Naikaku keeps the decision before and the record after.
Made for business automation, repeated research and investigation work, and code checks.
EMYSTI has published the way it works as noncommercial public source, so anyone can read it, fork it and contribute.
You want to use AI, but legal, data handling, quality and explaining things to customers make you careful. Naikaku keeps AI suggestions separate from actions and leaves a record people can check.
Run routine tasks, research, reports and code checks, only within what you allow, with a record kept.
See several AIs' opinions, critiques, retries and records on one screen.
Even when AI does the work, a person decides whether to accept it after looking at the log and the results.
EMYSTI Hero Series
We use Naikaku to keep planning, checking, translation, building and supervision separate when we work on VisitHero, TransHero, StayHero and HanasuHero.
A Japanese company using AI and languages to take the hassle out of daily life.
Field MapA multilingual app that answers "what now?" on a trip in Japan with a map and a next step.
CommunicationAn app that interprets conversations at hospitals, city hall and service counters.
Life in JapanA multilingual community where foreigners living in Japan can ask each other for help.
Everyday JapaneseAn app that lets you point at Japanese you see at home or school to translate it or hear it read aloud.
Start locally, check provider aliases and the gateway, then attach runner presets only when you need them. Raw API keys should stay out of browser storage.
Get the macOS developer preview from GitHub Releases.
2Use npm ci and npm run dev to open the Workbench.
Run npm run gateway for local provider and runner routes.
Configure provider, endpoint, model, and API-key alias per role.
5Allow only the CLI runners you choose, with evidence required.
Try it, connect APIs, supervise a Mac runner, or contribute. Each entry point keeps the same language structure.
Run steps, verification commands, and checksums.
Bring-your-own keys and future hosted gateway.
Boundaries for CLIs and Mac automation.
Pick the entry point by intent.
Contributions, adapters, and verification.
Noncommercial source and commercial permission.
Japanese comes first, but English and Chinese have the same content and the same page layout.
Same content.
Same content.
Same content.
A workbench where several AIs split the work: planning, doing, critique, supervision and scoring. What they do is kept on record.
Yes. Each role can use its own provider, endpoint, model, and API-key alias.
The public release is for noncommercial use. SaaS, client delivery, commercial modification, and paid integration require written permission from EMYSTI LLC.