We help secure the world’s most targeted organizations and products. We combine security research with an attacker mentality to reduce risk and fortify code.
October Trail of Bits Tribune: Releasing a PHP exploit demo, new plugins for patching, and package-integrity research. Plus Brotli-checker, SequenceHash, and six new public security reviews.
Read the Tribune: https://t.co/CIdlIFCCuQ
Patching is the hardest step for agents to get right.
We took what we learned from hundreds of Patch the Planet patches, thousands of internal Codex sessions, and a decade of fix reviews, and built it into two open-source agent plugins.
A good fix looks different in every codebase, and building one takes judgment and expertise. Our plugins encode best practices that work for our engineers so you can patch along at home.
•patching-bugs makes an agent prove each fix: the original code fails, the patched code passes, and a full revert fails again.
•regression-hunt checks what else the change broke, with a live proof of concept for every regression it reports.
Both work best with a THREAT_MODEL.md and patching guidelines, so agents know what your project considers a bug, a regression, and an acceptable fix.
They fit naturally in loops like @OpenAI's new Daybreak Defense Factory, where agents draft patches for people to review.
https://t.co/OjS56EB3pg
Announcing two new cyber products:
- Daybreak Defense Factory (https://t.co/OdlGbgPWp8)
- Daybreak Red Team (https://t.co/VQTAjDfduz)
Both make it much easier to put frontier cyber capabilities to work, helping defend companies in new ways.
Patch the Planet update: We’ve now merged 500+ fixes across 75 open-source projects, with 100 CVEs disclosed in the Linux kernel, curl, OpenSSL, nginx, vLLM, and more.
https://t.co/dDV7RUc56o
We're turning our lessons from defending OpenAI into practical tools for our customers. On Thursday, join us at Cyber at the Frontier, a live keynote, for our latest cyber announcements.
Working with our frontier models, we've identified and resolved important vulnerabilities in our own code and infrastructure. That experience is informing the tools and playbooks we're building to help other security teams strengthen their defenses.
@gdb shares what we've learned and why security teams should keep putting the latest models to work as capabilities improve.
Tune in on November 12 at 10 a.m. PT: https://t.co/jYly2SEDM0
My Dot found malicious code in an open PR on a @trailofbits Github repo. I didn't ask it to hunt for malware! It noticed a commit, inspected what changed, and warned me about an attack all on its own.
The biggest reduction in false positives we’ve seen in AI auditing came from giving the agent a clear threat model.
Put a THREAT_MODEL.md at the repo root and tell the agents to read it.
Cover:
• What the system does and how it’s deployed
• What needs protecting
• Who is trusted, and for which actions
• What an attacker can access or control
• Where untrusted input enters
• Which security rules must always hold
• What’s out of scope, and why
With the latest SOTA models, we’re seeing fewer hallucinations but more findings based on incorrect assumptions about the system, which is now the main source of false positives.
Codex Security��s threat-model skill is a good starting point, but the devs should own the document, verify every claim, and keep it up to date as the codebase evolves.
https://t.co/bIaM5gZQ5r
Patching has really become the hard part of our job while finding bugs is now one prompt away for anyone with Daybreak.
So why we turned our analysis of 130+ Patch the Planet patches into two skills to help devs and maintainers patch their code quickly and safely: https://t.co/qpjiJfIzuB
🚨Blockaid detected an ongoing exploit on @ankr on Ethereum.
An attacker submitted five withdraw calls on Ankr’s Ethereum cross-chain bridge, using a tiny Optimism deposit to unlock the bridge’s full locked inventory to a new wallet.
~$292k drained from the bridge so far across ~5 txs. Attack still ongoing. 🧵
Congratulations to the Cybersecurity Company of the Year finalists:
@Cloudflare | @DragosInc | @openssf | @shadowserver | @trailofbits
Vote for the community's choice by Nov 9.
Cast your vote: https://t.co/3eKsu0JjX3
Attackers can drain wallets by abusing a signing service, even without stealing private keys. Our 4-step post-Binance hack guidance helps wallet teams test what their signing systems can authorize and what limits the damage if signing is compromised.
https://t.co/ngd4FrIGHS
I find the “patch the planet” initiative quite inspiring. openai and others are spending huge sums of money on updating and securing all modern open source software infrastructure. it will be a continual effort to make the world more robust
Read the interview about how we approached our audits with @trailofbits@OpenZeppelin and @spearbit
Auditing such a novel software stack like Miden comes with interesting challenges. Trail of Bits has written some nice tooling even to make their lives easier
Our security researchers built an autonomous cyber vulnerability discovery, exploitation, and patching pipeline.
In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel at a total API cost of $1.2 million.
fresh fuzzy echidna, anyone? 🦔
Echidna 2.4.0 is out 🎉
🤖 MCP support for interactive fuzzing
🔬 considerably improved symexec via @argotorg hevm
⛓️ library autolinking with crytic-compile
🐛 fixes, reduced mem usage, and more
👉 https://t.co/R3ifmlbryc
🍺 also on brew by now
Very interesting work by the OpenAI Cyber team
Stronger models needing less orchestration is also something we quickly realized while building our main auditing harness at Trail of Bits (Auditron)
Instead of strong constraints from the harness, that were needed with previous models, we now focus on four things: using different model families, coordinating their work, enforcing persistence, and providing the right tools/skills at the right time rather than fixed recipes
A cryptographically verified tally doesn't mean every vote is legitimate. Paper ballots still matter: they provide an audit record independent of software. We explain how electronic voting works, and how cryptography can make paper-based elections more trustworthy. https://t.co/Hcnp8esM0M
Stable software often ships with known bugs because the fixes are never backported. That backlog is now an easy target for cyber-capable agents.
We used Codex to cut time per backport from days to ~30 min, with 149 Linux kernel CVE fixes applied so far.
https://t.co/p3AwFZtb3O
THORChain just inadvertently delayed the laundering of a significant amount of funds stolen from Binance.
The reason? An Ethereum reorg.
One happened this morning and caused THORChain to incorrectly issue a double refund to someone who had deposited 10 ETH.
This triggered an “emergency pause” on THORChain, preventing all users, including DPRK’s money launderers, from using cross-chain swaps.
Until the service comes back online, neither the hackers nor anyone else can swap through a majority of THORChain’s bridges.
We're aware a number of domains at https://t.co/4CEHuFgI4A were improperly redirected late Friday as part of a plan to access visitors' crypto wallets. The affected domains have been restored. We're investigating and will update this thread as we learn more.
Update following the Binance incident:
All markets are now unfrozen across Aave V3 and V4.
They were frozen as a precaution after initial reports, which paused new deposits and borrows while withdrawals and repayments continued as normal.
Aave contracts were unaffected by the incident and positions remained fully collateralized.
We identified a security incident affecting a modified Claude Code VS Code extension accessed through our Cursor IDE extension marketplace, impacting at most 5,000 users. This malicious version of the extension was not developed by Anthropic. It does not impact other extensions downloaded from the https://t.co/sJRGX5jLXW, Grok Bot, Grok or our other SpaceXSI products.
In response, we took down our extension marketplace this afternoon, worked with relevant providers to neutralize the malware, and have reinstated the marketplace after confirming there is no further risk.
If affected, please review instructions in this blog post to understand how to find the malicious package and establish a path to mitigate any potential impact. We are asking all customers update to Claude Code for VS Code version 2.1.314: https://t.co/ESqf7YEtHZ