Watson Standard1

← Back to the main page

Free written read

The short version

Free

Input
Ask with the form below. Once I've said I can take it, send your customer's security questions, a link to a public repo, or both.
Timing
I reply within 2 business days to say whether I can take it and when. Within 3 business days of having what I need, you get the read.
Output
A short written read: the three things most likely to stall the deal, each with why and the first step, and what I couldn't tell without your code.
Limits
I spend up to 2 hours on it, take two a week and do one per company. No call is needed, and you owe me nothing after it.
What it isn't
It isn't a filled-in questionnaire, and it isn't an audit, a penetration test, a certificate or legal advice. The full list.

A bigger customer has sent you security questions, or something else is about to ride on your software: a security review, a raise, a lot more users or a new AI feature. You'd like to know what's most likely to hold it up before you answer anyone.

You talk to the person doing the work: I'm Bryce Watson. I write the read with an AI tool helping, and I check every point myself before it goes to you. I spent more than a decade as a software engineer at eBay, and before that I worked on payments at hyperWALLET. See a sample report of how I read code. The free read is much shorter: three things.

When it fits

Not the fit: you need the questionnaire filled in, a certificate, or someone to test your running product. And if the answer depends on private code, that's the paid security first look, which reads it under an agreement.

What to send, and what I do with it

Please don't send private code, keys, passwords, customer data, or anything your agreements don't let you share. If something like that arrives, I delete it without using it and tell you. If a public repo has a key in it, I tell you, and I never use it. If you send material before I've said I can take it, I don't read it until I have, and the 3 business days start from my yes. If I can't take it, I delete what you sent within 7 days and tell you.

I use what you send only to write your read, and I don't contact your customer. I read it with an AI coding tool, Claude Code, which processes what it reads on Anthropic's servers; before I use it on what you send, I check that its model-training control is switched off. I don't share what you send with anyone else, and I decide what goes in the read. Within 7 days of sending the read, I delete my copies of what you sent (including the email it came in) and the session logs my AI tool keeps on my computer, and I tell you when it's done. Anthropic keeps its own records of what the tool read, under its own retention rules. I keep the read I sent you, which may quote from your questions, and the rest of our emails.

What you get

If I find fewer than three, I say so. I don't pad it.

What it isn't

How it works

  1. Tell me what you're running and what's coming, with the form below.
  2. I reply within 2 business days. If I can take it, I say when. I take two a week, so if both are taken I give you the next date before you send anything.
  3. You send the questions or the link by replying to my email.
  4. Within 3 business days of having them, I send the read. If I'm ill or something stops me, I tell you that day and give you a new date.
  5. Within 7 days after that, I delete my copies of what you sent, including the email it came in, and tell you.

Ask for a read

Nothing to pay and no access to give. I'll reply within 2 business days from bryce@watsonstandardco.com.

Don't paste the questions, keys, passwords or customer data into this form. It goes to my inbox through the Web3Forms relay, I use it only to reply, and I don't share it. What to send, and what I do with it.

or email bryce@watsonstandardco.com