←
Ron Perris637 posts
Ron Perris avatar

Ron Perris

@ronperris

Talks about secure coding for humans and AI.

◷ Joined March 2008▧ 39 Photos and videos
Jim Manico from Manicode Security avatar
⟳ Ron Perris retweeted
Jim Manico from Manicode Security@manicode·Oct 20

Claude Code and Codex are now writing a major portion of the world's new code. But why are they not talking about how secure the code is that they are writing, and instead are focusing on their hacking capabilities? Because AI does not code securely by default. There are dozens of reports and studies that echo this. Everyone is focusing on scanning that insecure code in all kinds of fantastic and clever ways! Manicode.ai is taking a very different route. We're literally defining a new AppSec/AI product category. We have built a massive trove of prompts and skills to use in the AI-based SDLC to get your agents to write secure code up front! It's super bad ass. But it's a very new product category that requires knowledge and expertise with AI based development. And we are here to help! jim@manicode.com #hitMeUp

Jim Manico from Manicode Security avatar
⟳ Ron Perris retweeted
Jim Manico from Manicode Security@manicode·Oct 11

Using AI to write code? Join me for a free webinar on making that code more secure. AI Secure Coding Maturity Model Oct 28, 8–9 AM Pacific / 11 AM–noon Eastern Clear requirements, useful context, tests, review, and a 30-day plan. manicode.com/webinar/

Jim Manico from Manicode Security avatar
⟳ Ron Perris retweeted
Jim Manico from Manicode Security@manicode·Aug 21

AI coding has changed my security workflow dramatically: Old way: Write code → review code → find security bug → fix bug → repeat. New way: Tell the AI the security requirements → generate secure code → verify it. Turns out the cheapest vulnerability to fix is the one your AI never writes. manicode.ai

Jim Manico from Manicode Security avatar
⟳ Ron Perris retweeted
Jim Manico from Manicode Security@manicode·Jun 29

Free webinar: Spec-Driven Secure Development with AI Thu, July 2 · 10 AM PT Vibe-coding ships bugs. Spec-driven coding ships secure software. In 60 min: - Bake security into the spec first - Drive Claude Code with secure coding prompts - Claude Code + Codex: build, review, verify - Live Q&A 100% demo 0 slides: us06web.zoom.us/meeting/regi…

Jim Manico from Manicode Security avatar
⟳ Ron Perris retweeted
Jim Manico from Manicode Security@manicode·Jun 26

The @OWASP_AISVS 1.0 is now released and live! github.com/OWASP/AISVS/blob/… cc @ottosulin @ricokomenda Russ Memisyazici and Raza Sharif

Ron Perris avatar
Ron Perris@ronperris·Mar 19

I felt a sense of relief today, knowing humans are not on the hook to solve so many problems alone anymore.

Jeremiah Grossman avatar
⟳ Ron Perris retweeted
Jeremiah Grossman@jeremiahg·28 Jul 2025

New startup has been founded! $12.5M seed round led by Ballistic Ventures. Vulnerability management — everyone knows it’s broken. With ~50% of breach losses originating for remotely exploited vulns, "fixing" this one area will have the greatest impact on cybersecurity. Thank you to everyone who supported us. It means the world.

We’re excited to announce Root Evidence, a company solving vulnerability management by focusing on the <1% that actually cause breach loss. Built by the team behind WhiteHat and Bit Discovery, Root Evidence is launching with $12.5M in funding led by Ballistic Ventures. Proof over probability. Fix what’s real. Read the full press release: rootevidence.com/news/root-e…

Post attachment
Jim Manico from Manicode Security avatar
⟳ Ron Perris retweeted
Jim Manico from Manicode Security@manicode·8 Apr 2025

Our course on rapidly building secure ReactJS applications with AI is now live! AI is horrible at secure coding. This course makes it dramatically better. udemy.com/course/the-complet…

ᴅᴀɴɪᴇʟ ᴍɪᴇssʟᴇʀ 🛡️ avatar
⟳ Ron Perris retweeted
ᴅᴀɴɪᴇʟ ᴍɪᴇssʟᴇʀ 🛡️@DanielMiessler·27 Jan 2025

One of the biggest impacts of AI that goes kind of unnoticed is that we’re about to see an explosion of poorly built applications. Specifically, applications built completely by AI with no thought of security whatsoever. 🧵

Post attachment