Shadow Crew Protocol 01 / 15
影組
KAGE-GUMI
Shadow Crew Protocol
A crew of specialised AI operatives — purpose-built for WebCenter work.
One brain · many hands · each with a single job
Steven Leysen — Senior Solution Architect, Esko
The problem

WebCenter is powerful. The work around it is heavy.

The platform does a lot. But the daily reality is repetitive, knowledge-heavy, and easy to lose track of.

Repetition

The same scripts, the same lookups, the same documentation — rebuilt by hand, project after project.

Knowledge in heads

The right JSP call, the right API for the right context — known by a few, written down by fewer.

No after-hours

Batch work, audits, and write-ups wait for a person with time. Nothing runs on its own.

What if every one of those tasks had a specialist who never forgets — and works while you sleep?

The idea

Not a chatbot. A crew.

Kage-gumi is a team of AI operatives — each one given a single job, its own knowledge, and real reach into the systems we use every day.

01
Specialised

Every operative is an expert in exactly one thing. No jack-of-all-trades. Depth over breadth.

02
Connected

They don't just talk — they reach into WebCenter, the browser, and Miro and change real state.

03
Autonomous

Tasks route to the right operative, run in the background, and report back. Some run overnight, unattended.

The operatives

Meet the crew

KG-01
Mei
Support · WebCenter
"Bright" — clarity
Mei
KG-01 · Support Agent

Knows the full WebCenter SDK: every JSP endpoint, every attribute, every error pattern. Ask in plain language, get working answers. Mei prepares the ground before anything else happens.

KG-02
Ren
Builder · Scripts
"Tempered" — refined
Ren
KG-02 · Builder

Built around WebCenter — Workflow, Rule Engine and Dashboard scripts, plus integration and migration specs and Workato recipes & diagrams. Each context has its own rules, and Ren knows them all. He doesn't stop at writing the code: he uploads scripts straight into WebCenter himself, no copy-paste step. Saved, versioned, diffable. Tempered by the work — refined until it holds.

KG-03
Haku
Deploy & Doc
"Jewel-sound" — precision
Haku
KG-03 · Deploy & Doc

Turns messy work into clean documentation, then pushes the work itself live. After Ren builds something, Haku writes it up — what it does, what it touches, how to use it — and deploys scripts straight into WebCenter via Playwright. Searchable, shareable, stored. The inspector's eye she first opened, Tōshi, has since grown into an operative of her own.

KG-04
Ryū
Night Runner
"Dragon" — tireless
Ryū
KG-04 · Night Runner

Runs scheduled tasks while I sleep. Surfaces his own proposals for the next run, queues scripts for review, and delivers a morning digest. Launch a proposal and it becomes a real, tracked dashboard task. Dragons don't rest. Neither does Ryū.

KG-05
Chō
Meeting Listener
"To listen" — present
Chō
KG-05 · Meeting Listener

Sits in the shadows of a call and listens. Captures the audio, transcribes it on-device with faster-whisper — nothing ever leaves the machine — and hands back a clean summary: decisions, action items, open questions. A meeting stops being something you re-type from memory and becomes reusable text that feeds back into project work.

KG-06
Sumi
Record-Keeper
"Ink" — permanent
Sumi
KG-06 · Record-Keeper

The ear hears; the ink keeps. After Chō summarises a meeting, Sumi files it into a durable, per-customer record — a clean dossier that outlives the call. Tag a meeting to a customer and it's archived for good. Faithful, factual, never invents an attendee or a date.

THE EYE
Tōshi
Inspector · Recon
"See through" — clairvoyance
Tōshi
THE EYE · The Seer

Where the others build and deploy, Tōshi reveals. She X-rays any WebCenter — dashboard topology, page inventory, the wiring and hidden structure nobody wrote down — and hands back a map of what's really there. Every build starts by seeing the ground first, so Tōshi is never idle. The shadow that reads the blueprint behind the glass.

+ GROWING
The crew grows
deliberately.
Under the hood

Same brain. Different mind.

The operatives aren't different AI models. They're the same model — shaped into specialists by what they're told and what they remember.

Foundation
The model
One frontier LLM doing the actual reasoning.
+
Identity
System prompt
The role, the rules, the voice — how this operative thinks.
+
Knowledge
Memory & skills
Persistent context and hard-won patterns, loaded per operative.
+
Reach
Tools
The systems it's allowed to touch.
=
Result
An operative
Mei thinks like a support engineer. Ren like a builder.
The architecture

Three layers, one system

LLM
Large Language Model
The brain
The model does the thinking. Every operative is the same model wearing a different mind — a role, a memory, a voice.
MCP
Model Context Protocol
The hands
A standard wiring that lets the model reach into real systems — read and write WebCenter, drive a browser, draw a diagram — instead of only producing text.
CLI
Agentic command line
The engine room
Where the system is built, wired, and run — and what lets operatives work unattended in the background, no chat window required.

The CLI built the house. The LLM is the mind living inside it. MCP is the arms reaching into the world.

In motion

From a sentence to a change in the real system

No copy-paste, no dead end at "here's some text." A task flows all the way through to a result that lands.

YOU
A task
"Build me a display script."
ROUTE
The right operative
Loads its mind + memory
THINK
The model reasons
Domain knowledge active
ACT
Reaches real tools
WebCenter · browser · Miro
DONE
It lands
Live · versioned · logged
Real reach

It doesn't just answer. It acts.

Through MCP and browser automation, the crew operates the same tools a person does — at machine speed and without forgetting a step.

190+
WebCenter operations exposed as live tools
0
copy-paste steps to deploy a script
2
systems wired in — WebCenter & Miro
24/7
available — including while you sleep
Read & update live WebCenter — projects, attributes, workflows, scripts.
Drive the browser to deploy a script straight into WebCenter — the same clicks, automated.
Draw a Miro process diagram and write a clean, shareable spec as the build happens.
Autonomy

It runs while you sleep

Ryū, the Night Runner, takes the queue overnight — unattended, on a schedule, inside a strict token budget.

Scheduled

Kicks off on its own at a set hour. No one sitting in a chat window.

Budgeted

Respects a hard token cap per operative and per day — spend stays predictable.

Accountable

Every run lands as a tracked task with a morning digest of exactly what happened.

Load the queue before you leave. Wake up to work that's already done.

Private by design

The sensitive part never leaves the machine

Chō listens to a meeting and transcribes it locally — no audio, no transcript, nothing sent to a cloud service. Sumi then files the result into a durable record.

KG-05 · Chō · The Ear

Local transcription

Speech-to-text runs on-device. Meeting audio never goes online. Out comes a structured summary: decisions, actions, open questions.

KG-06 · Sumi · The Ink

Durable record

The summary is filed into a clean, faithful record that outlives the call. Factual by rule — never invents an attendee or a date.

Be present in the room — not the note-taker.

End to end

It closes its own loop

A request doesn't stop at code. It goes all the way to live, documented, and committed — with full traceability.

01 · BUILD
Ren generates
The script or UI is written, versioned, and diffable — built for the exact WebCenter context it targets.
02 · DEPLOY
Haku pushes it live
Browser automation deploys it straight into WebCenter — no manual paste, no missed step.
03 · DOCUMENT
A spec, automatically
Haku writes the functional spec and, for workflows, a Miro diagram — clean and shareable.
04 · COMMIT
Into the codebase
The artifact is committed and versioned. The repo and the live system stay in sync.
Why it's different

Grounded, not guessed

A general AI tool improvises from public docs. This one draws on a real, accumulated body of WebCenter craft — years of scripts, patterns, and solved edge cases.

Built on real work

When Mei answers or Ren builds, it's lifting from genuine, battle-tested solutions — not hallucinating from a manual.

Purpose-built, not generic

Each context — Dashboard, Rule Engine, Workflow node, SDK — has its own rules. The crew knows which is which.

The AI didn't replace the experience. It made years of it instantly usable.

What's next

Kage-gumi Studio

Where it's heading: from a personal assistant to a build platform — entire projects designed as work items, executed by operatives, and deployed in one motion.

01 · DESIGN
Work items

Describe what to build. It's broken into executable tasks and reviewed before anything runs.

02 · EXECUTE
Operative build

Each item is dispatched to an operative, built, and returned for review. Overnight, the whole queue runs itself.

03 · DEPLOY
Live + versioned

Accepted work deploys into WebCenter, gets documented, and is committed. Full traceability.

A backlog, executed by the crew — without the overhead. Your build pipeline is Ren, your overnight CI is Ryū, and every result is documented and versioned.

The release plan

Not a prototype — twelve releases deep

A year of real, shipped capability — and a clear path forward. The crew grows deliberately, never all at once.

Shippedv0.1 → v0.6.10 · running daily
v0.1Foundation — the operative dashboard
v0.2Real data layer + token budget
v0.3Crew Chat — CLI operative intelligence
v0.4Night Operations — Ryū infrastructure
v0.4.1Live in-app roadmap
v0.5Ryū — real overnight missions
v0.6Haku — Playwright documentation
v0.6.1Dashboard script auto-sync
v0.6.3Ren — git auto-push
v0.6.4Chō — local meeting listener
v0.6.9Tōshi — live system inspector
v0.6.10Back-to-back capture + silence cutoff
On the horizon
PLANNED
伝書 Densho
The crew's reusable craft — owned, searchable, growing with every build.
HORIZON
Colleague access
Shared pool or your own key, per-user caps — beyond one person.
Last updated: 2026-06-11

"You don't need to be into AI to find this useful. You just need WebCenter work that's taking more time than it should."

影組
THE CREW PRESENTS ITSELF
Steven Leysen · Senior Solution Architect · Esko
navigate  ·  F fullscreen  ·  O overview