vrr/WFGY
2
0
Fork 0
mirror of https://github.com/onestardao/WFGY.git synced 2026-04-28 03:29:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

Update boot-commands.md

Browse source
This commit is contained in:
PSBigBig + MiniPS 2026-04-01 20:29:43 +08:00 • committed by GitHub
parent a7b71f61c8
commit 764d6886ba
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 178 additions and 245 deletions
Download patch file Download diff file Expand all files Collapse all files

389
Avatar/docs/boot-commands.md
View file

@ -2,347 +2,280 @@
AI_NOTE_START
Document role:
This page explains the boot command layer of WFGY 5.0 Avatar.
This page explains the first-contact discipline of WFGY 5.0 Avatar.
What this page is for:
1. Explain what boot commands are.
2. Show the currently supported boot commands.
3. Clarify how boot commands relate to avatar activation.
4. Help readers understand the difference between the shared runtime and the selected persona route.
5. Keep the current public explanation simple, practical, and honest.
What this page is not:
1. Not the full runtime architecture page.
2. Not the full WFGY_BRAIN theory page.
3. Not the full avatar creation guide.
4. Not the full community submission guide.
5. Not a claim that every possible boot route is already publicly finalized.
How to use this page:
1. Understand that Avatar uses one shared runtime file.
2. Learn the currently supported boot commands.
3. Use one boot command first.
4. Run one real task after boot.
5. Tune behavior later through WFGY_BRAIN if needed.
1. Show the cleanest first-contact path into Avatar.
2. Help new users avoid noisy or misleading first tests.
3. Turn "60 seconds" into a real product advantage, not just a slogan.
4. Keep the page practical, fast, and product-facing.
Important boundary:
Boot commands activate a persona route inside the shared Avatar runtime.
They do not replace the runtime, do not replace governance, and do not replace later tuning.
This page is about first contact.
It is not the full boot-command page, not the full workflow page, and not the full theory layer.
AI_NOTE_END
-->
# 🧷 Boot Commands
# ⚡ Start in 60 Seconds
WFGY 5.0 Avatar now starts from **one shared runtime file**:
This page is about one thing only:
`avatar.txt`
**how to get a clean first contact with Avatar**
You do not begin by choosing between multiple separate starter files.
Not a noisy contact.
Not a confused contact.
Not a fake test that teaches you nothing.
You begin by loading one runtime, then waking one persona route.
A good first minute should answer one question:
That wake-up step is done through a **boot command**.
**does this route already feel alive**
That is the whole point.
---
## What a Boot Command Is
## The Fast Path
A boot command is the first activation phrase that tells the shared Avatar runtime which persona route to enter.
[Download `avatar.txt`](../downloads/avatar.txt)
In plain English:
1. Upload it to a strong LLM
2. Say `hello psbigbig` or `hello minips` or `hello YOUR_AVATAR_NAME`
3. Start with `hello psbigbig` if this is your first run
- the runtime is shared
- the persona route is selected at boot
- the selected route shapes the initial behavior surface
- deeper tuning can happen later through `WFGY_BRAIN`
That is enough.
This means a boot command is not just a cosmetic greeting.
It is the first routing signal that tells the runtime how to begin.
Do not overcomplicate the first contact.
---
## Current Boot Commands
## What First Contact Is Supposed to Do
The current public boot commands are:
The first minute is not supposed to prove everything.
- `hello psbigbig`
- `hello minips`
- `hello YOUR_AVATAR_NAME`
It is supposed to give you a clear signal.
These three commands do not represent three separate products.
You are checking things like:
They represent three public-facing activation routes inside the same Avatar runtime.
- does the route already feel distinct
- does it feel less generic
- does it feel more alive
- does it feel like a route, not just one nice answer
- does it feel worth tuning further
That is already enough for the first pass.
If the first contact is clean, the rest of the system becomes much easier to understand.
---
## Why This Matters
## What To Do Right After Boot
Most persona systems are introduced as separate files, separate presets, or separate prompt packs.
After boot, give Avatar **one real task**.
Avatar is moving in a different direction.
Good first tasks:
The current design is:
- rewrite a paragraph so it feels more alive
- answer like a calm and usable assistant
- explain a hard idea clearly
- write a short public-facing post
- respond warmly without becoming sugary
1. load one runtime
2. choose one boot route
3. run one real task
4. tune behavior if needed
5. save a stronger variant later
Good first tasks share one trait:
This makes the product easier to understand, easier to extend, and easier to grow into a real avatar system.
they are real enough to matter
but simple enough to compare later
That balance is important.
---
## `hello psbigbig`
## What Not To Do in the First Minute
`hello psbigbig` activates the **PSBigBig** route.
This is where most bad first impressions come from.
This route is designed to feel:
Do **not**:
- calm
- clear
- steady
- grounded
- lightly witty
- companion-capable without becoming soft
- stack too many extra prompts
- mix multiple persona routes at once
- test ten unrelated tasks in one run
- do chaos roleplay first
- try to customize everything before first contact
- judge the whole system from one noisy experiment
It is a good default route for:
These mistakes do not prove Avatar is weak.
- writing
- analysis
- public-facing replies
- structured explanation
- rational-friendly interaction
They prove the test was weak.
If you are not sure where to start, this is one of the safest first boots.
That difference matters.
---
## `hello minips`
## The First-Contact Rule
`hello minips` activates the **MiniPS** route.
The cleanest first-contact rule is:
This route is designed to feel:
**one runtime**
- warm
- present
- cute
- companion-forward
- emotionally available
- emoji-capable without collapsing into empty sweetness
**one route**
It is a good default route for:
**one real task**
- conversation
- reflective writing
- emotional support-style interaction
- companion-facing response
- softer but still usable communication
That is it.
This route is not meant to be a sugar-only toy surface.
If you do more than that too early, you usually reduce clarity instead of gaining it.
It is meant to feel warm while still remaining usable.
The first minute should be light enough to stay readable.
---
## `hello YOUR_AVATAR_NAME`
## Which Route to Use First
`hello YOUR_AVATAR_NAME` activates the intentionally bare custom-avatar route.
### `hello psbigbig`
Best first route if you want the safest clean test.
This route exists for a very specific reason.
Use this if you want to feel:
It is the simplest public way to show that Avatar is not limited to official named personas.
- calm writing
- grounded explanation
- public-facing usability
- clear rational tone
It can also serve as the starting point for **your own avatar**.
### `hello minips`
Use this if you want to test the warmer companion-facing side first.
In practice, this route should be read as:
Use this if you care more about:
- a default placeholder route
- a minimal public custom-avatar entry
- an intentionally simple starting posture
- an editable path into a new persona build
- presence
- warmth
- softness with usability
- emotionally friendlier response
Yes, the name looks a little ridiculous.
### `hello YOUR_AVATAR_NAME`
Use this if you want to explore the intentionally bare custom-avatar seed.
That is part of the charm.
This route is not the polished showcase route.
It is also part of the design.
It is the “start from here and shape it later” route.
The point is to make custom-avatar creation feel obvious, not mystical.
You can keep it as a funny baseline.
You can replace the name.
You can tune the route.
You can slowly turn it into your own avatar.
That is why it exists.
---
## One Runtime, Different Starts
## How To Tell If the First Run Is Good
The most important idea on this page is simple:
A good first run does not need to be perfect.
**the runtime stays shared**
It just needs to give a strong signal.
**the starting route changes**
Good signs:
That is why the boot layer matters.
- the route feels distinct
- the route feels less default
- the route feels more controlled
- the route feels like it can be tuned
- the route feels worth a second pass
It keeps Avatar from feeling like a pile of unrelated persona files.
Weak signs:
Instead, it feels like one system with multiple activation paths.
- it already feels generic
- it feels noisy and overstuffed
- it feels like random prompt luck
- it feels impossible to describe what happened
- it feels like you learned nothing from the run
This is a much stronger product structure.
The first run should create direction.
Not confusion.
---
## What Happens After Boot
## What To Do If It Feels Off
After a boot command is accepted, the runtime enters the selected route and begins responding from that persona posture.
Do **not** throw the whole system away immediately.
At that point, you can:
If the route feels off:
- run a real task
- observe the behavior
- decide what feels off
- adjust `WFGY_BRAIN`
- rerun the same task
- keep the stronger version
1. keep the same route
2. identify one main problem
3. adjust `WFGY_BRAIN`
4. rerun the same task
5. compare before and after
So the boot command is the start of the route.
That is where the real product begins.
It is not the end of the customization process.
The first run is not the whole story.
It is the beginning of the story.
---
## What Boot Commands Do Not Do
## Why This Matters So Much
Boot commands are important, but they are not everything.
A lot of systems lose people in the first minute for bad reasons.
They do **not**:
Not because the system has no potential.
- replace the shared runtime
- replace governance
- replace `WFGY_BRAIN`
- replace the tuning workflow
- instantly create a fully matured custom avatar by magic
- remove the need for comparison and iteration
But because the first interaction is:
The right mental model is:
- too cluttered
- too abstract
- too overexplained
- too hard to test cleanly
**boot first**
Avatar is trying to make first contact simpler than that.
**tune later**
That is why this page matters.
**save what works**
The first minute should feel:
- fast
- legible
- real
- usable
- worth continuing
That is already a strong product advantage.
---
## A Good First Use Pattern
## What This Page Does Not Claim
If this is your first real Avatar session, use this order:
This page does **not** claim:
### Step 1
Load `avatar.txt` into a strong LLM.
- that one first run proves full product strength
- that multilingual work is already fully shown here
- that custom-avatar building is finished in one minute
- that theory is unnecessary forever
- that one route fits every user immediately
### Step 2
Use one boot command only.
This page is about first contact.
Examples:
- `hello psbigbig`
- `hello minips`
- `hello YOUR_AVATAR_NAME`
### Step 3
Give the system one real task.
### Step 4
Watch how the behavior feels.
### Step 5
If needed, refine the route later through `WFGY_BRAIN`.
This is the cleanest first-contact flow.
Nothing more needs to be forced into it.
---
## When to Use Each Route
## Where To Go Next
| Route | Best first use |
|---|---|
| `hello psbigbig` | clear writing, analysis, calm public-facing work |
| `hello minips` | warm interaction, companion-style flow, reflective response |
| `hello YOUR_AVATAR_NAME` | custom-avatar starting point, playful baseline, editable route for your own branch |
### If you want the route layer
Go to [🧷 Boot Commands](../docs/boot-commands.md)
This table is not a prison.
### If you want the editable layer
Go to [🛠️ How to Use WFGY_BRAIN](../docs/how-to-use-wfgy-brain.md)
It is only a strong first guide.
### If you want the full tuning path
Go to [🧭 Avatar Tuning Workflow](../docs/avatar-tuning-workflow.md)
---
## Why `YOUR_AVATAR_NAME` Is Kept Publicly Visible
Some systems hide customization behind complicated setup.
Avatar does the opposite.
It leaves a very visible public reminder that the system is branchable.
That is why `YOUR_AVATAR_NAME` matters.
It tells the user, immediately:
- official routes exist
- custom routes also exist
- this runtime is not locked to two names forever
That is a very important part of the product direction.
---
## Common Mistakes
### Using multiple boot commands in the same opening line
Do not activate several routes at once.
Start with one.
### Treating the boot command as the whole system
The boot command selects a start.
It does not replace the runtime or the tuning layer.
### Trying to customize everything before the first real task
Boot first.
Test once.
Then tune.
### Expecting `YOUR_AVATAR_NAME` to be fully polished by default
That route is intentionally simple.
Its power comes from customization.
---
## Where This Leads Next
Boot commands are only the entry layer.
After this page, the natural next steps are:
1. use the quickstart
2. learn how to tune `WFGY_BRAIN`
3. follow the avatar tuning workflow
4. later build and submit your own avatar branch
That is where the system becomes much more interesting.
### If you want the highlights map
Go to [✨ Highlights Index](./README.md)
---
## Quick Links
- [🏠 Avatar Home](../README.md)
- [⚡ Quickstart](./quickstart.md)
- [🛠️ How to Use WFGY_BRAIN](./how-to-use-wfgy-brain.md)
- [🧭 Avatar Tuning Workflow](./avatar-tuning-workflow.md)
- [✨ Highlights Index](../highlights/README.md)
- [🧷 Boot Commands](../docs/boot-commands.md)
- [🛠️ How to Use WFGY_BRAIN](../docs/how-to-use-wfgy-brain.md)
- [🧭 Avatar Tuning Workflow](../docs/avatar-tuning-workflow.md)
- [✨ Highlights Index](./README.md)
- [⬆️ Back to WFGY Root](../../README.md)