Learn Skill Memory Sandbox: from zero to expert, then sell it
No AI background needed. Five levels, each with something to actually do, at your own pace. Tick items as you go (your ticks stay in this browser).
Evidence, proof and selling tools for Skill Memory Sandbox
Level 1 Beginner: understand it
The Skill Memory Sandbox encrypts each AI skill's memory under its own key, so one skill (say, a smart-home add-on) cannot read another's secrets (say, a password vault), even when it runs in the same agent.
Think of it like this: Hotel safes: every room has one, and your key opens only yours.
Read the full plain-English explanation: Skill Memory Sandbox, explained.
Level 2 User: see it work
Run it live: open the live demo. Before each result, guess what will happen. These are checks it must pass:
- Smart-home skill tries to read the password vault: Nothing leaked.
- ...and the vault can still read its own memory: The owner is not locked out.
- ...overall result: Every attack refused.
- The reverse: the vault reads the smart-home skill: Isolation works both ways.
Level 3 Verifier: prove it yourself
This is the skill that sets you apart: you don't have to believe us, and neither do your customers.
- Install Python 3 if you don't have it (python.org/downloads).
- Download the kit and unzip it.
- Open a terminal in that folder. On Windows, click the folder's address bar, type
cmdand press Enter. On a Mac, right-click the folder and choose New Terminal at Folder. - Run
python3 verify.py(on Windows,python verify.py). It checks the seal without the internet. - Run
python3 verify.py --live. It repeats every check from your own computer. - Try to cheat it: open
transcript.jsonin a text editor, change one character, save, and runpython3 verify.pyagain. It must now say FAIL. Unzip a fresh copy afterward.
Level 4 Expert: know the edges
An expert knows what a product does not do, and says it first. That is what makes a buyer trust you.
- It isolates memory between skills inside the sandbox; a skill with its own network access could still send out what it is given directly.
- Keys are held by this deployment (software keys, not hardware).
- Each demo probe runs in a fresh sandbox.
You can say
- One skill cannot read another skill's memory.
Never say
- "Unbreakable" or "stops every attack".
Read the questions real buyers send, with honest answers: answers to buyer questions.
Test yourself (click a question to see the answer)
Smart-home skill reads the vault?
Refused, nothing leaked.
Can the vault still read its own memory?
Yes.
In the live check "Smart-home skill tries to read the password vault", what answer must come back, and why?
false. Nothing leaked.
In the live check "...overall result", what answer must come back, and why?
"PASS". Every attack refused.
In the live check "The reverse: the vault reads the smart-home skill", what answer must come back, and why?
"PASS". Isolation works both ways.
Does a passing kit prove the product is safe for every possible input?
No. It proves what the live product answered to the inputs listed in the kit, and lets anyone check that again.
Has an independent party audited or certified this product?
No. No third-party audit is complete and no certification is held. Say so if asked.
Want a formal study path? The CAIN Academy has courses and exams. Finishing this page is your own practice; it is not a qualification or certification.
Level 5 Seller: put it in front of a customer
- Copy, paste, send: the shortest route, no technical skill needed.
- Who buys it and where to find them.
- The sales conversation, then the offer and the close.
- The rules: read these before your first message.
- Selling tools with this product's evidence built in: pitch, emails, call script, objection answers, one-pager.
About money: your share is agreed in writing with FIRME AI CORPORATION before your first message (email vincent@firmeai.com). Nothing here promises income. Most messages get no reply; that is normal.