
#LangSec conspirator eleminating injection vulnerabilities, #DevSecOps practitioner authoring #SecurityBelts and #ThreatModeling security architecture.
Joined January 2012
- Tweets2K
- Following307
- Followers207
- Likes2.7K
Pinned Tweet
Summing up the root cause of input-based security vulnerabilities dev.to/bob5ec/the-root-cause…
Lars retweeted
CVE-2026-98432 is a good example of something that really concerns me about the future of cybersecurity.
ubuntu.com/security/CVE-2026…
The vulnerability was introduced in October 2015, more than 11 years ago, by a performance optimization in the Linux TCP/IPv4 networking stack. One of the most fundamental components of an operating system used by millions of servers.
Now Anthropic reports that Claude discovered this flaw and developed a working exploit that allows remote code execution as root, without authentication, against a server with an MPTCP listener.
anthropic.com/research/disco…
Think about what that means.
For decades, we’ve been building critical infrastructure out of enormous amounts of complex, memory-unsafe code. Humans wrote it. Other humans reviewed it, tested it and searched for vulnerabilities. All working with the same fundamental limitations: limited time, limited attention, incomplete visibility and an inability to reason about every possible interaction and race condition.
And perhaps that’s what protected us to some extent. Attackers had the same limitations as the people developing and defending these systems.
A vulnerability could remain undiscovered for 11 years, even in such a widely used component.
Now AI is changing that balance.
The ability to inspect code, identify subtle weaknesses and develop working exploits is becoming less dependent on a small number of highly skilled researchers. These activities can increasingly be scaled with compute, automated testing and tokens.
And that’s what worries me about the next few years.
We’ve accumulated decades of complex, memory-unsafe code in operating systems, network services, applications and libraries. Much of it has never been assessed with anything close to the capabilities that are now becoming available.
We may have massively overestimated the security of our software simply because the people looking for vulnerabilities were as limited as the people who introduced them.
That balance is breaking down, and I don’t think we’re prepared for what follows.
Lars retweeted
I'm releasing my personal security utility prompts in waves and licensing them under MIT.
* Free Threat Modeling Prompts: github.com/jmanico/grim-prom…
* Free Third Party Library Reduction Prompts: github.com/jmanico/trim-reap…
* Free Security Logging Expansion Prompts: github.com/jmanico/dread-led…
I hope this shows you the value of what we are building here!
For teams looking to guide secure coding from the start, manicode.ai offers commercial secure-coding prompts for AI assistants, designed to reduce avoidable security rework and the time and tokens spent on it.
Lars retweeted
I conjecture that no LLM, even with an infinite amount of tokens, can find an efficient attack on AES, BLAKE3, ChaCha, or SHA3, for no such attack exists (also, P!=NP)
[ with infinite tokens/time it could theoretically break these algorithms, but not find a non-trivial attack, where a trivial attack is "hash messages A and B that I found to collide" ]
Lars retweeted
🤘🏻 talk with @endrazine at @unpromptedconf
slides aumasson.jp/data/talks/unpro…
to appear in @phrack soon
Lars retweeted
🛠️ Fuzzing confused dependencies with Depfuzzer
New tool designed to automate the detection of dependency confusion vulnerabilities
Repo: github.com/synacktiv/DepFuzz…
By @Synacktiv
synacktiv.com/en/publication…
Lars retweeted
🦀 Eliminating Memory Safety Vulnerabilities at the Source
Rust caused memory safety vulnerabilities % in Android to drop from 76% to 24% over 6 years.
💡Key insight: new code is disproportionately responsible for bugs
By @jeffvanderstoep, @ayper
security.googleblog.com/2024…
I had a great time meeting the @OwaspSAMM Community. Lots of like-minded people!
TIL: OWASP SAMM is for people running an #AppSec program. Target groups are not Developers or Security Champions. Hence OWASP DSOMM and Security Belts need to be there to support these target groups.
Greetings from Lisbon. Looking forward to @OwaspSAMM User day and @owasp AppSec Global.
48
Lars retweeted
Join me for FREE, live #AppSec Training to celebrate the launch of @Semgrep Academy! academy.semgrep.dev 🚀
🔒 Building an Application Security Program
June 20: Level 3
Register: ow.ly/puGP50RgpiC
Observe the ones around you. Do they have True Ownership? blog.alexewerlof.com/p/broke…
Lars retweeted
There should be a reality show where project managers try to meet outrageous deadlines while developers keep introducing new features.
Looking forward to talk about Security Coaching at @heise_devSec with @jn_struewer and @SpreadTheKaozZ heise-devsec.de/veranstaltun…
1
69
For everyone who is improving security culture, ui-patterns.com/ might be an awesome source of inspiration for fundamental patterns that can be applied.
44
Lars retweeted
🤖 Building an AI AppSec Team
Using @crewAIInc to create a multi-agent AppSec team
* Code reviewer
* Exploiter
* Mitigation expert
* Report writer
#cybersecurity
srajangupta.substack.com/p/b…
Lars retweeted
I just launched a new post with @clintgibler over on tl;drsec, check it out!
When I read Wiring the Winning Organization (@RealGeneKim, @StevenJSpear), I spent the whole time trying to map the concepts to Security
1/2
Looking forward to see you all at Global AppSec and @OwaspSAMM User day in Lisbon!
Ready to raise your #cybersecurity game? You don't want to miss Global AppSec Lisbon 2024! Join us June 24-28 for incredible training and talks in inspiring Lisbon, Portugal.
Get your tickets now at lisbon.globalappsec.org/
46
Lars retweeted
We just published an almost complete list of talks that have been accepted for #TROOPERS24. Thanks to all of you who participated in the CFP! So many excellent submissions. We really had a hard time to decide which will fit best for this year!
troopers.de/troopers24/talks…
Lars retweeted
An underrated aspect of AppSec and Secure Coding is not exposing the insecure functionality in the first place.
Let's say you have a XML parsing library that may be used by devs wrongly/insecurely. By disabling certain functions in the library, its not vulnerable to XML Injection anymore
Instead of constantly training them to figure out security params, having a wrapper library (custom) that automatically disables insecure functionality is way more effective.
It's:
* easier to use
* easier to enforce (in SAST, CI, SBOM, etc)
* easier to train on
* reduces cognitive load for devs in the long run
* and more secure
Keep it simple.
























