omashiki: work delegated to AI agents, with control and review
Open-source platform that takes tasks, runs each one with AI agents in an isolated environment, and returns a result ready for review.
- Service
- AI agents, Process automation
- Area
- IT
- Repository
- github.com/This-Is-NPC/omashiki ↗
01The problem
Routine tasks, like fixing an issue, updating several repositories, or putting together a report, are either done by hand or handed to AI tools with no control over access, secrets, or traceability.
How the solution works
-
01
Receive taskEngineering teams · delegate tasks
-
02
Queue
-
03
Run isolatedAI agents · do the work
-
04
Check secrets
-
05
Hand over for reviewReviewers · accept the result
02What we did
We built an execution layer that takes tasks from trackers, schedules, forms, or other agents, puts each one in a durable queue, and runs it in an isolated environment. The result comes back in a reviewable form, such as a branch, a verified file bundle, or a completion record, with a leaked-secret check and a signed notice when it finishes.
03What it changes
Engineering and operations teams can delegate routine work to AI agents knowing what each one accessed, and review everything before accepting it. It works from a laptop to a shared pool of machines.
More cases
An AI assistant that runs inside the company
Employees talk to the company's documents without anything leaving its servers, and each person only sees what they are allowed to see.
View case →Mori: factory cameras that warn you when something is off
Uses the security cameras already in place to detect risky situations and alert the team within seconds.
View case →omunculus: teams of AI agents with rules and oversight
Open-source engine that coordinates several AI agents: it hands out work, limits what each one may do, calls a person when needed, and records every step.
View case →