Python Runtime Boundary for Jay

AI-era Runtime Development · Not Traditional Root Server Deployment

Your Runtime: jayho80.homelinked.tech


1. Important Message

You do not need sudo for most Python Runtime development.

Many online tutorials use sudo because they assume you own the whole VPS or cloud server.

In our HSIMC Runtime system, your role is different.

Traditional VPS tutorial:
You control the whole server
↓
sudo install
sudo nginx
sudo systemctl
sudo certbot

HSIMC Student Runtime:
You control your own Runtime workspace
↓
Python
SQLite
JSONL
HTML
Evidence
Exports
Admin publishes trusted URLs

This is an AI-era Runtime model.

You are not expected to manage the entire server. You are expected to create capability, evidence, working code, and URL-level outputs inside your own Runtime.


2. Your Workspace

Your main working directory is:

/opt/workspaces/jayho80

Your website directory is:

/var/www/jayho80

You can work inside these directories.

Use:

cd /opt/workspaces/jayho80
source venv/bin/activate

Then you can run Python and install Python packages without sudo.


3. What You Can Do Without sudo

Example:

cd /opt/workspaces/jayho80
source venv/bin/activate

pip install flask requests pandas openpyxl

mkdir -p app exports sqlite jsonl logs

No sudo is required for this.


4. Your Standard Python Test

Create a simple Python test:

cd /opt/workspaces/jayho80
source venv/bin/activate

cat > app/hello.py <<'PY'
print("Hello from Jay Runtime")
print("AI-to-Code → Code-to-URL → URL-to-Evidence")
PY

python app/hello.py

If it prints the message, your Python Runtime is working.


5. SQLite Test

Create a local SQLite database:

cd /opt/workspaces/jayho80

python3 - <<'PY'
import sqlite3
from pathlib import Path

db_path = Path("sqlite/jay_runtime.db")
db_path.parent.mkdir(parents=True, exist_ok=True)

conn = sqlite3.connect(db_path)
conn.execute("""
CREATE TABLE IF NOT EXISTS evidence (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    title TEXT,
    status TEXT,
    created_at TEXT DEFAULT CURRENT_TIMESTAMP
)
""")
conn.execute("INSERT INTO evidence (title, status) VALUES (?, ?)", ("Jay Runtime SQLite Test", "ok"))
conn.commit()

for row in conn.execute("SELECT id, title, status, created_at FROM evidence"):
    print(row)

conn.close()
PY

This also does not require sudo.


6. JSONL Evidence Test

Write one JSONL evidence record:

cd /opt/workspaces/jayho80

python3 - <<'PY'
import json, datetime
from pathlib import Path

p = Path("jsonl/events.jsonl")
p.parent.mkdir(parents=True, exist_ok=True)

record = {
    "event": "runtime_test",
    "student": "jayho80",
    "status": "ok",
    "message": "Python Runtime works without sudo",
    "ts": datetime.datetime.utcnow().isoformat() + "Z"
}

with p.open("a", encoding="utf-8") as f:
    f.write(json.dumps(record, ensure_ascii=False) + "\n")

print("written:", p)
PY

7. How to Create a Webpage

Create your page inside exports:

cd /opt/workspaces/jayho80

cat > exports/demo.html <<'HTML'
<h1>Jay Runtime Demo</h1>

<p>This page was created inside Jay's Runtime without sudo.</p>

<pre>
AI-to-Code
↓
Code-to-URL
↓
URL-to-Evidence
↓
Evidence-to-Offer
</pre>

<p>This is the correct AI-era Runtime workflow.</p>
HTML

Then tell Admin:

Please publish:
/opt/workspaces/jayho80/exports/demo.html

to:

https://jayho80.homelinked.tech/demo/

Admin will publish it using the trusted publishing tool.


8. What Admin Handles

The following actions require Admin:

This protects the main HSIMC Runtime server and other students' Runtime spaces.


9. Why This Is Different from Online Tutorials

Most online tutorials assume:

one student
one VPS
one root admin
one website

Our Runtime system is different:

many students
one trusted Runtime node
separate workspaces
separate evidence
separate URLs
Admin-controlled publication
shared safety boundary

Therefore, copying a normal VPS tutorial directly may be wrong.

The goal is not to become a root server administrator first.

The goal is to become an AI-assisted URL-level Capability Builder.


10. Your AI-era Workflow

Observe a real problem
↓
Use AI to generate code
↓
Run Python inside your Runtime
↓
Create SQLite / JSONL evidence
↓
Generate an HTML page
↓
Submit exports to Admin
↓
Admin publishes trusted URL
↓
URL becomes Evidence / Offer / Capability Link

This is the correct workflow for Capability Commerce.


11. Migration from Old Python System

If you already have Python code on your local computer or from an old server, migrate it into this structure:

Old Python code
→ /opt/workspaces/jayho80/app

Old HTML pages
→ /opt/workspaces/jayho80/exports

Old SQLite databases
→ /opt/workspaces/jayho80/sqlite

Old logs
→ /opt/workspaces/jayho80/logs or /opt/workspaces/jayho80/jsonl

Old documents
→ /opt/workspaces/jayho80/docs

Old evidence
→ /opt/workspaces/jayho80/evidence

Then activate your Python environment:

cd /opt/workspaces/jayho80
source venv/bin/activate
pip install -r requirements.txt

If there is no requirements.txt, install only what your project needs.


12. Boundary Principle

Student owns capability creation.
Admin owns trusted publication.

中文:

学生拥有能力创造。
管理员拥有可信发布。

This is not a limitation. It is the structure that makes a multi-student Runtime safe, trustworthy and scalable.


13. Final Reminder

Do not ask first:

How do I get sudo?

Ask first:

Can I build this inside my Runtime workspace?
Can I run it with Python venv?
Can I output HTML / SQLite / JSONL?
Can I prepare an Evidence URL?
Can Admin publish the trusted URL?

Most student Runtime work does not need sudo.

Use AI to write code. Use your Runtime to test. Use Evidence to prove. Use URL to exchange capability.

中文版说明

这份中文说明解释 Jay 的 Python Runtime 权限边界、迁移方式,以及为什么大多数学生 Runtime 工作不需要 sudo。

打开中文版 / Open Chinese Version