Pinned Tweet
I've released CrownGuard, a free MIT-licensed tool that finds the crown jewels in your Microsoft 365 or Google Workspace, scores their risk and builds a branded PDF report. It runs in your browser. I'm particularly proud of this one. github.com/jusso-dev/crowngu…
108
Open source I starred this week.
AWS has a local emulator of its APIs, so you can test code without an AWS account: github.com/aws/aws-emulator
shadcn's improve skill has your best model audit a repo and write plans for cheaper models to build: github.com/shadcn/improve
1
25
ServerOS panel runs the Linux servers you already own from one web app, with deploys, backups, uptime checks and logs: github.com/serveroshq/panel
Ember is Slack-style chat that runs entirely in your own Cloudflare account: github.com/NYLLON-SOFTWARE/e…
0 is an open source AI security harness for code analysis and vuln research. Bring your own model: github.com/0sec-labs/0
GARM auto-scales ephemeral self-hosted runners for GitHub Actions and Gitea across clouds and Kubernetes: github.com/cloudbase/garm
10
Just as I suspected. Microsoft Threat Intelligence saw phishing exploiting Microsoft 365's outage. Emails impersonate service health notices and claim held mail will be deleted in 24 hours unless you click to restore.
Don't click. Check the admin centre.
xitter.cf/msftsecintel/status/21…
Microsoft Threat Intelligence observed an active phishing campaign exploiting the Microsoft 365 service disruption on November 12. The messages impersonate Microsoft 365 service health notifications and reference the legitimate Exchange Online incident EX1550543.
Some of the phishing emails Microsoft detected on November 13 falsely claim that messages were held during the outage and will be deleted within 24 hours unless recipients click to restore them.
Observed subjects follow the pattern "Outage - [six-character code]," with sender display names using organization-specific prefixes followed by "Outage Notification" or "Service Health."
The campaign uses three sender domains, moments-notice[.]com, planetmaison[.]com, and newbooker[.]com, with recipient-specific SendGrid click-tracking links redirecting to reallifeaccounting[.]com, including the observed path /themelist0.html. Similar messages targeted organizations across multiple US sectors.
The activity is consistent with credential phishing, although credential collection and malware delivery have not been confirmed.
Security teams can use Microsoft Defender for Office 365 Threat Explorer and Microsoft Defender XDR advanced hunting (EmailEvents, EmailUrlInfo, and UrlClickEvents) to identify suspicious messages, examine embedded URLs, and investigate user clicks.
Microsoft Defender for Office 365 Safe Links provides time-of-click protection against malicious URLs when properly configured. Organizations should verify their Safe Links policies, investigate potential user interactions, and confirm service disruptions directly through the Microsoft 365 admin center.
55
After this week's Microsoft 365 outage, I'd expect more Microsoft support scams: calls, emails, texts or pop-ups about refunds, reverify requests, MFA, remote tools or password resets.
Microsoft won't cold-call you. Open the status page or admin centre yourself. Report it to IT.
53
Microsoft has a new Entra preview called Hard Deletion Protection. A protected user, group, app or service principal can't be permanently deleted for 7 days after a soft delete. One catch: once it's on for an object, you can't turn it off.
learn.microsoft.com/en-us/en…
69
ASD's ACSC has published guidance on securing AI testing environments. It points to cases from the UK AI Security Institute, OpenAI and Anthropic where test agents reached real systems outside the intended scope. Prompts aren't containment.
cyber.gov.au/publication/sec…
7
Railway's breach report is worth reading. A build escape leaked a credential their Temporal cluster trusted. A bad OIDC binding did the rest. The target was AI keys: up to 100k workspaces, ~$740k stolen spend.
blog.railway.com/p/incident-…
10
Anthropic replaced Git. OpenAI kept it and rebuilt the server, and median push latency fell nearly 80%. GitHub's Rubicon took 7,200 pushes a second from 5,000 agents in testing.
Same problem: agents write code faster than source control can take it. Review and CI are next.
1
14
The ACSC hasn't issued new open-weight guidance. The 5 November publications are still current. If a model is already running, I'd start with a register of what's deployed, controls that sit outside the model, and a rollback plan.
cyber.gov.au/publication/ope…
13
2-8 Nov (UTC): 3,347 unique vulns, down 3.4%. 12-week avg 3,504 vs 941 a year ago. YTD 94,600 vs 41,021. KEV added 3; 5 new Metasploit modules. Oct keyword match: 604 AI-related vulns vs 81. 24 Oct CVEs said AI helped find the bug (230 YTD, a floor). Reported counts, not causes.
6
Justin Middler retweeted
We just shipped AWS Emulator, which gives coding agents a local copy of AWS to test against.
Early in my career, we built test environments like this by hand, one system at a time. Now, agents are writing code far faster than any team can check it in production. Matt Goldberg, a Senior Principal Engineer on our Developer Tools team, wanted to know whether agents could build and test those environments themselves. He gave @kirodotdev a single prompt, and about 15 minutes later it came back with a first version of an SQS emulator. Turing that into a robust solution took much more work
Here’s what I think this means for engineers today: allthingsdistributed.com/202…
Qwen4-Max is in preview on Alibaba's Token Plan, as qwen4-max-preview, until 25 November (UTC+8). Pay-as-you-go isn't available, and the docs say the model will be retired when the preview ends.
docs.qwencloud.com/developer…
78
Microsoft has released .NET 11. It's Standard Term Support, so you get two years, and .NET 10 remains the LTS option. C# 15 arrives with union types. If you're still on .NET 8 or 9, support ended on 10 November.
devblogs.microsoft.com/dotne…
6
Microsoft is retiring its SMS and voice options for Entra password reset from 1 February 2027. A passkey for sign-in doesn't automatically count as a recovery method. Check who still relies on them, and give those users another way to reset a password.
entra.news/p/entra-news-174-…
14
Cisco and IBM are putting Anthropic's Mythos bug-finding model into their security products. You get the capability, not the model. Cisco says it's coming soon for select enterprises, against a registered, authorised target. IBM is aiming for early 2027.
theregister.com/ai-and-ml/20…
10
I added the ASD Essential Eight and ISM suite to Maester. It's 14 checks for MFA, privileged access, Office macros and patching, and it's in the 3.0 preview. You can run it against your Microsoft 365 tenant alongside CIS and CISA.
maester.dev/docs/tests/asd/
12
I'm building Bilby, a free app that will tell non-technical people, in plain English, whether their computer looks healthy. The window already opens, but nothing checks the machine yet. Windows first, and I wouldn't treat it as a real scan today.
github.com/jusso-dev/Bilby
9
Justin Middler retweeted
Microsoft has released its November security updates. This update included:
• 1,235 vulnerabilities patched.
• 1 vulnerability with evidence of exploitation.
• 66 'Critical' rated.
Users should apply the patch updates ASAP 👉msrc.microsoft.com/update-gu…
Security updates for November 2026 are now available. Details are here: msft.it/6018SZEg0
#PatchTuesday #SecurityUpdateGuide
Anthropic has released Claude Fable 5.2 for long-horizon agent work, at $8/$40 per million tokens, down from 5.1. It's on the API, Bedrock, Google Cloud and Microsoft Foundry. It needs 30-day data retention, and Fable will block most cyber tasks.
anthropic.com/claude-fable-a…
33







