DOGE worker’s code supports NLRB whistleblower
krebsonsecurity.com
krebsonsecurity.com
Original code: https://github.com/Ge0rg3/requests-ip-rotator
Forked: https://github.com/markoelez/async-ip-rotator
Code is pretty much the same, with comments removed, some `async` sprinkled in and minor changes (I bet this was just pasted into LLM with prompt to make it async, but if that worked why not).
Except... Original GPL3 license is gone. Obviously not something you would expect DOGE people to understand or respect.
Archived repo page: https://archive.ph/LI7tt; archived previous repo count: https://archive.ph/tgkg5
0. https://arstechnica.com/tech-policy/2025/04/i-no-longer-hack...
You can download it as a Git repository from https://archive.softwareheritage.org/api/1/vault/git-bare/sw...
I say "may", because I'm not sure if you have internal code on a public git or FTP server, is that consider "distributing"?
“If this were a side project, it would just be bad code,” the reviewer wrote. “But if this is representative of how you build production systems, then there are much larger concerns. This implementation is fundamentally broken, and if anything similar to this is deployed in an environment handling sensitive data, it should be audited immediately.”
Look at the critique [0] and then look at the code [1].
[0] https://web.archive.org/web/20250423135719/https://github.co...
[1] https://github.com/ricci/async-ip-rotator/blob/master/src/as...
Certainly Elon made him print it out on paper to personally code review.
x_forwarded_for = headers.get("X-Forwarded-For")
if x_forwarded_for is None:
x_forwarded_for = ipaddress.IPv4Address._string_from_ip_int(
randint(0, MAX_IPV4)
)
lolIt's possible, but very unlikely, the copyright license wasn't actually violated because, for example, the fork could have arranged a separate license.
The best example of this is the Qt Project's code: https://www.qt.io/qt-licensing
You can get it under a GPL license for free. You can pay them money to get it under a Commercial license that would let you modify the code without releasing changes.
So, while I doubt it happened, the person who forked it here could have contacted the original author, the copyright holder, and asked for an exemption from the GPL terms.
You are only required to keep the GPL3 license if you re-distribute it. Putting it in a GitHub repo, is ambiguous whether or not it is re-distributing it, at least morally.
If you want to delete the license in a personal copy, that is perfectly valid according to the license terms. If you then happen to upload that to a private GitHub repo, also perfectly valid.
If you then happen to upload that to a public GitHub repo, because of, say, restrictions on free private repos, without intent to distribute, then what?
Then you keep the license eh? Distributing without an intent to distribute is distributing.
Git is free and open source. If you want version control and collaboration and NO unintended distribution completely for free you can just use Git. It even has a built in server to share with your work buddies.
Vibe coding
> Original GPL3 license is gone. Obviously not something you would expect DOGE people to understand or respect.
Why would they? They don't give a FF about courts.
To be fair I see in my daily life folks who copy and paste from stack overflow or random GitHub repo and move on with their day. They ignore the Creative Commons Attribution-ShareAlike or whatever license is applied to the code they copied.
I see on this very site people who will share copyrighted articles that are behind a paywall (just because it is on some archive site doesn’t make it right).
Please don’t take this as support for DOGE and the headaches they are causing. To make a cheap jab at a group of people while ignoring the group that you associate with is bad form.
Copy pasting from stack overflow without attribution is wrong but it's also harder to claim "ownership" over single lines or small snippets. It depends how "obvious" they are. You definitely can't copyright trivial functions. There's a lot of gray here but yes attribution is always good.
But things get a lot less murky when we're talking about forking a project. That's usually nontrivial and non obvious. I think what's most important is that removing a license is an active decision. Certainly that would make a critical difference in a court [0]
Then there's further escalation by who is doing the action. The more power and influence you have the greater responsibilities. All men are not created equal. Men with more power can disproportionally do more damage and require higher accountability. So yeah, I care a fuck ton more about a government employee doing something bad especially while performing official duties more than some rando. The ability to do harm is very different.
The reason I dislike your comment is because it's dismissive of the action. "Other people do it!" Is not a defense nor excuse. It is even worse by ignoring multiple points of context.
[0] though protecting open source has been traditionally hard for many reasons. Specifically it's hard for small developers to take legal action, especially against larger bodies. But isn't this something we should want to be fixed? Credit for our own contributions?!
What group does the person who makes the comment associate with?
irony
" On or about March 11, 2025, NxGen metrics indicated abnormal usage at points the prior week. I saw way above baseline response times, and resource utilization showed increased network output above anywhere it had been historically – as far back as I could look. I noted that this lined up closely with the data out event. I also notice increased logins blocked by access policy due to those log-ins being out of the country. For example: In the days after DOGE accessed NLRB’s systems, we noticed a user with an IP address in Primorskiy Krai, Russia started trying to log in. Those attempts were blocked, but they were especially alarming. Whoever was attempting to log in was using one of the newly created accounts that were used in the other DOGE related activities and it appeared they had the correct username and password due to the authentication flow only stopping them due to our no-out-of-country logins policy activating. There were more than 20 such attempts, and what is particularly concerning is that many of these login attempts occurred within 15 minutes of the accounts being created by DOGE engineers. "
The worst possible interpretation is straightforward - they are working for the Russians as agents and let the Russians in or installed the keyloggers for Russia.
Excerpt: "How much more proof do we need that this administration is completely compromised? There is zero reason for the US to relax any offensive digital actions against Russia. If anything, we should be applying more."
They were accessing Github over the internet from superuser accounts they were presumably also using as their user account. Given the code quality, I doubt their opsec is put together, either.
The objective may not have been to obtain access or any useful data. The objective may have been to get the scary headlines about Russians and use the existing media and political agitprop to further destabilize the government you seek to color revolution away.
If they're trying to exfiltrate data, they might want to rotate through IP addresses in order to obfuscate what's going on or otherwise circumvent restrictions. Using a simple ip rotator like the post talks about would maybe be an approach they'd use. If they're not careful with the IP addresses, once in a while one might get caught due to some restriction like being outside the US. It'd maybe appear as though you're getting these weird requests from Russia, but that's just because you're not logging the requests that are not being flagged from the US.
Maybe I'm reading the post incorrectly though (if so, please correct me!)
Best possible case I see would be that the whistleblower has made some mistake (or is being intentionally dishonest). Seems plausible for instance that "it appeared they had the correct username and password" based on "our no-out-of-country logins policy activating" could just be a misunderstanding of how/when the policy triggers. Not to say it's the most likely explanation, just the least concerning one.
I think less concerning than keyloggers, while still assuming the whistleblower is correct, would be that a DOGE employee was using a VPN/proxy/Tor. Probably not a great idea to have traffic going through a hostile nation state even with encryption, but less bad than keyloggers on their machines stealing and trying credentials within minutes.
Definitely concerning though, to be clear - just steelmanning/answering the question of best possible interpretation.
Though with nation state actors you can't rule out Pegasus like zero-click infiltrations.
This is the evidence which strongly suggests that the DOGE personnel are using various cloud IP addresses to scrape.
This context enables two important things:
- Granular exceptions: If Alice is attending a conference in Toronto, you can say "Allow Alice to log in from Canada next week" without opening Canada-wide logins for everyone. Pre-auth geo-blocking forces you into an all-or-nothing stance.
- Better threat intelligence: A valid login from an unexpected region (e.g. Moscow when Alice is normally in D.C.) is a far stronger signal of compromise than a failed attempt. Capturing "successful login + wrong location" helps you prioritize real threats. If you block pre-auth, you'd never know Alice's account was compromised.
Putting geo-checks after authentication gives you precise control over whom, exactly, is logging in from where, and offers richer data for your security monitoring.
[1] https://learn.microsoft.com/en-us/entra/identity/conditional...
Why would they attempt a login from Russia (if it was indeed Russians)?
It is incredibly cheap to use a VPN with a US residential IP.
Many non technical people use VPNs to access region restricted content. It is trivial to understand and use.
Assuming this all actually happened as described, it sounds like someone wanted it to appear that these attempts were coming from Russia.
Probably the least expected location to connect from, if it was genuine. Not saying it necessarily isn't, but it's not usual either and doesn't make much sense.
Feels like a pretty good Occam’s razor case… but is there any legitimate reason why one would request this?
DOGE needs to be shutdown and everyone of them held as a flight risk while the whole thing is investigated.
I never thought I'd be calling for UN observers for an election in the US but here we are
DOGE is a complete clusterfuck. Fwiw I think there is hard to spot fraud in the govt that should be looked at (eg price inflation at the pentagon, VA, Medicaid/Medicare, SS). They should have done the hard work of uncovering that. Instead they just went for clickbait headlines.
https://whistlebloweraid.org/wp-content/uploads/2025/04/2025... - page 2 & 11
"This declaration details DOGE activity within NLRB, the exfiltration of data from NLRB systems, and – concerningly – near real-time access by users in Russia. Notably, within minutes of DOGE personnel creating user accounts in NLRB systems, on multiple occasions someone or something within Russia attempted to login using all of the valid credentials (eg. Usernames/Passwords)"
"For example: In the days after DOGE accessed NLRB’s systems, we noticed a user with an IP address in Primorskiy Krai, Russia started trying to log in. Those attempts were blocked, but they were especially alarming. Whoever was attempting to log in was using one of the newly created accounts that were used in the other DOGE related activities and it appeared they had the correct username and password due to the authentication flow only stopping them due to our no-out-of-country logins policy activating. There were more than 20 such attempts, and what is particularly concerning is that many of these login attempts occurred within 15 minutes of the accounts being created by DOGE engineers."
https://krebsonsecurity.com/2025/04/whistleblower-doge-sipho...
https://www.npr.org/2025/04/15/nx-s1-5355896/doge-nlrb-elon-...
> Within minutes after DOGE accessed the NLRB's systems, someone with an IP address in Russia started trying to log in, according to Berulis' disclosure.
https://www.npr.org/2025/04/15/nx-s1-5355896/doge-nlrb-elon-...
Is this normal to build this sort of functionality into a software system? Especially software systems that heavily rely on auditability?
My company retains all e-mails for at least 5 years, for audit purposes. But if some troublemaker were to e-mail child porn to an employee, we'd need to remove that from the audit records, because the laws against possessing child porn don't have an exception for corporate audit records.
So there's essentially always some account with the power to erase things from the audit records.
"No" is the answer to GP: there is no legitimate reason for a fully unlogged superuser account.
From the previous post, they had auditor roles built in that they purposely chose to go around
You always need it to setup the system initially.
It's like root on Linux: it's an implementation detail that it must be possible.
There is no legitimate justification for this request.
Anything musk's dogs claim to find cannot be taken at face value because of this. Because there is no audit, and no evidence that they can offer that they didn't doctor their findings.
The next time they claim that a 170-year old person is receiving SS checks, they have no way to prove that they didn't subtract a century from that person's birthdate in some table.
That statement might be (slightly) more believable had there not been access attempts from Russian IP addresses using valid (and recently created) DOGE login credentials so very shortly thereafter.
They want to prove that AI can do "just as good a job" on these data sets and arrive at "equal conclusions" with a much higher level of effiency.
This is what happens when you get high on your own supply.
Of course, given the blatant dishonesty and criminality that the rest of this administration is producing (see: every immigration law case that they are losing in court), you'd have to be a useful idiot to actually assume good intent from them.
I think if I wanted to describe an account with access to perform "sudo -s" as negatively as possible, I would say "an all-powerful admin account that is exempt from logging activity that would otherwise keep a detailed record of all actions taken by those accounts."
https://www.pbs.org/newshour/show/nlrb-whistleblower-claims-...
...
While he was at work, and it also contained photographs of him walking his dog taken by a drone.
This is mafia shit.
That's the best I could do. LOL
Some previous attempts for DOGE to get data has resulted in data being deleted before they can look and requests for judges to block access to data.
DOGE may be trying to be covert in order to stop these two activities from happening before they can get and review the data.
By definition, a judge decides what's legitimate.
If DOGE expects their access to be blocked by a court judgement, and bum-rushes agencies to exfiltrate data ahead of the judgement, that's also criminal intent.
I am not sure what you are getting at. "Covert" isn't how I'd describe DOGE's actions. "Brazen" maybe?
What’s happening with judges is very political. We likely won’t know what’s allowed until things have gone through the appeals process. There have been cases of judges admitting they will rule against the current administration no matter the topic or law. This is messy, to say the least.
This fairly clear.
The story says that DOGE attained access to an account that had huge permissions into what it could see and alter. The person or persons from DOGE may have downloaded 10GB of data. The person may have used this in a manner that is illegal. Or it is illegal to start with. With the understanding that POTUS may or may not be allowed grand such access. (I dont think POTUS can)
2. DOGE employee downloaded code that could be used to use a huge pool of IP addresses, from AWS to bypass forms of throtheling. 3. The code was badly written. 4. The person is a racist
How would a person from DOGE use "unlimited" number of IP adderssess from AWS to hammer and automaticlay screenscape webpage, benefit from it when it came to copying extremly sensetive data from an internal National Labor Relations Board database?
Did 10.000 sessions authenticate to the database at the same time, using AWS UP addresses and scraped the data?
Something is pretty broken if the system with extremly sensetive data is available from external IPs -and- allowing a single account to login 10.0000 times to concurrently scrape data off the interal database?
Of are they saying that this code was adapted to use 10.000/100 IP addresses internal to National Labor Relations Board and scrapes using those?
The automation later noted makes a lot more sense to aid the work.
What data in a federal agency could the chief executive not have authorization to access?
We can also add to that IRS data. The articles of impeachment against Nixon included the following:
"He has, acting personally and through his subordinates and agents, endeavoured to obtain from the Internal Revenue Service, in violation of the constitutional rights of citizens, confidential information contained in income tax returns for purposes not authorized by law" (emphasis mine).
There actually are laws regulating the handling of personal data collected by the government and it generally doesn't have a "the president wants to see it" exception.
I wonder, if he was alive today would he stand by those words?
I think that he can access a health or irs record for cause - anything which would not get him impeached.
Personally? For starters, he can't access anything the Legislature's laws say he can't.
The Executive is there to implement the law, and that includes obeying them him/her-self.
A President telling other people to break the law on his behalf by threatening to fire them is also a crime of extortion.
Can the legislature make rules for the president without constitutional amendment?
I am interested- I’ll see if I can find examples.
That said, I was surprised to learn much later that, by all accounts, Elon Musk was a competent and resourceful leader in SpaceX's early days. Maybe these stories are just his personality cult in action, but I found it plausible. It appears he once knew his place as an engineering manager, without LARPing as a Chief Engineer (he didn't appoint himself to CTO until quite a bit later). I worked for a really good manager who didn't know how to code, but he knew a lot about software and was very good about pulling back on coding things vs pushing forward on software design. It seemed like Musk was similar at SpaceX.
Which is all to say that celebrity is a helluva drug. I don't think Musk was ever an especially "high-IQ individual," and his first marriage suggests he's always been a misogynistic loser. But being anointed "a real life Tony Stark!" seems to have destroyed his brain. Ketamine probably doesn't help.
He's good at having and raising money which was what SpaceX needed, I think he was probably the same then as he is now. Reading about his early days at Tesla and the PayPal stuff, I don't really buy the idea he was ever different and took a dark turn. He's the type of person that will never self-regulate and somehow has never faced any negative consequences for lying and self-aggrandizing so has kept pushing it further
Eric Berger's book in particular suggests that, before Falcon 1 was successful, Musk was much more humble and collaborative with the other early SpaceX hires, and typically deferred to their expertise. He was always reckless and megalomaniacal. But after Falcon 1 he became much worse.
I mean, I guess this really happens in all industries. Art, music, leadership, software development. People who maybe once had credibility in something and now desperately try to foist Their People as the best in the industry.
I feel like that is what is happening here. None of the people who Elon surrounds himself are notable in any way, and their skills are hugely suspect, but he has to have his harem of "Super Coders" to prop up his own mythology.
(Btw, awesome username!)
Not trying to defend the means to the end, but I would really like my tax money used more efficiently. I will also say am extremely worried about the levels of access that they are being given, especially since it comes with basically no accountability
If you could prove that billions were saved in pure waste, then I’d imagine any sane citizen would agree with you, setting aside matters of decorum and human decency (e.g. RIFs that may ultimately be necessary but conducted in an inhumane way)
I’d like my tax money used efficiently, but this group does not merit the trust to carry out those changes, even on a technical level
That is what the GAO is for https://www.gao.gov/ , and these people are much better than script kiddies.
> I would really like my tax money used more efficiently
Me too! You are on hacker news so I assume you are firm believer in https://en.wikipedia.org/wiki/Amdahl%27s_law ! If you would like your tax money used efficiently, are you willing to discuss cuts to social security, medicare, medicaid, veteran benefits, and whatever else is at the top of the budget? https://www.cbo.gov/publication/61181? What would you cut?
Personally, I would increase taxes on anyone making over $500K/year and stop nickle and diming our federal government so the US can actually become a first world country for everyone that isn't a software engineer.
Then a lot of those had to be reinstated because you simply can't operate a hospital without sanitation.
Just like they had to scramble to hire back the folks at the National Nuclear Safety Association.
Yeah, efficiency is great. But this is like ordering tacos and getting... a used tire and some dirty diapers...?
This is like the derelict father with partial custody who parachutes in one weekend a month to buy his son ice cream and a new video game to leave two days later the conquering hero. Meanwhile mom works two jobs, has to set all the expectations and responsibilities for the child, and the father is late on child support payments.
DOGE blitzkrieged government IT. It'll be years before we understand the scope of what they've done and given available evidence: these are script kiddies who worship Musk, I don't think there is ANY reason for optimism or charitable consideration.
Except by most accounts so far it was being used efficiently by the federal workforce. This whole debacle will end up costing the US tax payer more money. See cutting the IRS or USAID which will probably lead the US to bailing out farmers. And if they privatize, then it'll be even more expensive.
This is immature thinking, because, who wouldn't?
The contention comes from differing opinions on what is waste.
I agree, but for a different point.
Generalising, but under the age of 25, most people don't have enough experience (business/government) to understand things such as business ethics, the consequences, auditing practises, privacy concerns, etc.
With professional experience, you develop a better understanding and build up that depth of knowledge of how things impact the wider "world" rather than the immediate task at hand. Meaning, you gain a better understanding of the ethical implications of what you're doing.
As an example - in law, it'd be easier to manipulate a law graduate than a lawyer with 20+ years experience, who would think outside the direct question or task that was asked.
Why is the age such a talking point? An insight in startup culture is that the public underestimates young people, especially when it comes to business skills with objective results and tight feedback cycles.
It just seems like now that HN is skewed older we no longer believe that?
I also think your description applies well to many startup founders.
One could be a 60 year old skid.
Is the federal government a business or startup? Does chainsawing it have a tight-enough feedback cycle to get good results? I'm going to say No to both, but I don't have the time or expertise to try to prove it. It can be true, both that young people are great at startups and bad at the federal govt.
This is supporting evidence that they were picked for ideological reasons - namely, being young white supremacists who wouldn't ask any questions. That's why Marko Elez was rehired. Musk, Trump, and Vance all share his views, he was just dumb enough to express them in public.
Young people get convinced to do stupid things that their older selves would regret.
> On February 6, someone posted a lengthy and detailed critique of Elez’s code on the GitHub “issues” page for async-ip-rotator, calling it “insecure, unscalable and a fundamental engineering failure.”
Link from quote: https://github.com/markoelez/async-ip-rotator/issues/1
The follow comment is interesting to be a coincidental, such a weird interaction.
I must be missing something here; surely the level of elite technical skill implicit in his résumé would preclude this kind of thing
I would say that Elmo picked a bunch of junior devs because they don't have enough maturity to talk back and will do anything they're asked but I think that's too charitable. I think he actually went this route because Elmo is a sad man in his 50s who is desperately trying to pretend that he is, and has not matured beyond, his 20s.
Musk did a "poll" on X that voted for rehiring Elez to DOGE, by February 20th Elez had a US Government email address again, and on Febrary 21st he was reported as working for DOGE at the Social Security Administration.
the 2nd comment in the issue explains why the 1st was posted pretty clearly
The public repos for this person that I could find that weren't forks with no activity to upstream consisted of a dice-rolling guessing game, rock-paper-scissors, and some kind of framework for downloading and transcribing audio files that does not yet download or transcribe, but implements a whole bunch of boilerplate. I find it rather difficult to believe this person engaged in a good-faith review of the async-ip-rotator code base.
Go look at the list of pardons this administration has handed out. These guys won’t even be charged.
That seems like a lot. Source?
Just as its only worth complaining about geriatric geezers in office until the cheeto man brings in young hackers, then the problem is that "the old impaired people were good, actually".
Don't observe. Don't think. Merely repeat the approved message.
> The Party told you to ignore the evidence of your eyes and ears. It was their final, most essential command.
This is the deep state they've been worried about, this is the boot that will tread on them.
EDIT: parent comment was highest ranked comment for the article and is now at the bottom?
We live in a nation of laws, whether or not conspiracy-minded individuals prefer to follow them.
You stopped living in a nation of laws a while ago. Now you live in a nation of might makes right.
One of the things that is being exposed by the current administration is that, even though the Judiciary is an arm of the government, and supposed to provide a check on the Executive, the reality is that the Executive has the power to pardon anyone it sees fit, voiding the power of the judiciary (the argument is that the ultimate power lies with the voters who can pass their judgement on the Executive, and its use of its powers, by voting them out, hopefully)
[1]: https://constitution.congress.gov/browse/essay/artII-S2-C1-3...
[2]: https://constitution.congress.gov/browse/essay/artV-1/ALDE_0...
[1]: https://www.newspapers.com/article/news-and-record-truman-ex...
If this is all true, this is basically hacking sensitive data in the open. We already know the current administration has worked to hobble unions. So putting these things together, this act is not only wrong in and of itself, but the data is likely going to be used to harm americans' interests. So, deserving of punishment.
And then the means to do so have involved ignoring the courts and bypassing constitutional checks and balances? Please tell me how this isn’t criminal if not treasonous?
After all, why do they need unfettered access? Why do they need your bank statements? Why do they need to hide what they're doing with the unfettered access?
That's what's happening here. There is no good explanation other than bad actors
If Joe Blow off the street walks into a federal agency and takes all their data – open and shut case, throw the book at them, see you in a few decades.
If someone from the White House walks into a federal agency, tells the agency leadership "the President wants me to take all your data", and the agency leadership replies "sure, go right ahead" – not a scenario people were expecting, so the existing laws haven't been crafted to clearly criminalize it. Maybe some enterprising prosecutor can find a way to map it to the crimes on the statute book, maybe it is just too hard. But even if the prosecutor overcomes that hurdle, it will be far from easy to convince the jury / trial judge / appellate courts that the legal elements of the crime are actually met – and if it actually gets as far as a conviction upheld by the appellate court, what do you think the conservative SCOTUS majority are going to do with that when they get it? And many prosecutors, foreseeing those low odds of ultimate success, will stop before they even get to an indictment.
So, I think the odds of anyone ultimately being convicted over this are low, even if Trump never pardons them.
Maybe, Congress might pass a law to make it more clearly illegal, which might make it easier to prosecute if a future administration repeats the same behavior.
EDIT: if people are downvoting this because they think my analysis of the likelihood of successful criminal prosecution is wrong, it would be great if they could reply to explain where they think I got it wrong
If you are going to charge them with a crime, which one? CFAA?
How then to prove that access is unauthorized under the CFAA given evidence that both the President and senior agency leadership authorized it? Trying to claim that those authorizations are legally invalid gets into rather murky areas of law, and is (AFAIK) without precedent. Can you point to any previous cases of a successful CFAA prosecution where the access was authorized by a senior federal official but that authorization was declared legally void?
How do you get past the fact that the law is ultimately whatever SCOTUS says it is, and it seems more likely than not that the majority of current SCOTUS will want to say that this specific situation isn't a crime?
I feel like people are rejecting my position because they don't like it or don't want it to be true. Of course, maybe I'm wrong – maybe Thomas, Alito, Gorsuch, Kavanagh, Barrett and Roberts are all secretly dreaming of sending Musk and his minions to federal prison; or maybe they'll dispassionately follow their own judicial philosophies to the logical conclusion that doing so (using CFAA or whatever) is statutorily and constitutionally required - but that doesn't seem very likely to me, given their track records. Do you really think I'm wrong about that?
If you don't feel that way then you deserve the world you are creating.
So, how do you prosecute them for accessing a computer system (or data or whatever) without authorization when both the President and the senior agency leadership say they authorized it?
Well, you can’t-unless you want to argue that the President / agency leadership’s authorization is illegal and hence illegally invalid, ultra vires. But even supposing you are right about that in the abstract, will you be able to convince a judge and jury of it? And even supposing you convince a jury, trial judge and appellate court, there’s a dozen different ways SCOTUS could overturn it (from narrow questions of statutory construction to sweeping rulings about the President’s inherent constitutional power to demand information from the executive branch), and I think the main question for the current SCOTUS majority will be which of those ways they choose.
My impression is that a lot of people are mixing up what they think the law ought to be, with what it actually is. Just because something ought to be a crime doesn’t mean it actually is one - and that’s especially going to be the case with unprecedented situations, it is hard to make something a crime if nobody foresaw it would one day happen.
Is this some reminder to people that bad things occur that aren't found out.
Considering how everyone is aware of this is your comment some sort of clusterbomb whataboutism?
Suggest reading the complaint: https://whistlebloweraid.org/wp-content/uploads/2025/04/2025...
https://whistlebloweraid.org/wp-content/uploads/2025/04/2025...
> Furthermore, on Monday, April 7, 2025, while my client and my team were preparing this disclosure, someone physically taped a threatening note to Mr. Berulis’ home door with photographs – taken via a drone – of him walking in his neighborhood. The threatening note made clear reference to this very disclosure he was preparing for you
A little nit-picking, but that's not what open source means, especially as it relates to the GPL in this case. If you can't use the code commercially, it's neither "open source" (as defined by OSI) nor free software (as defined by the FSF).
"Tim Bearese, the NLRB's acting press secretary, denied that the agency granted DOGE access to its systems and said DOGE had not requested access to the agency's systems. Bearese said the agency conducted an investigation after Berulis raised his concerns but "determined that no breach of agency systems occurred."
https://www.npr.org/2025/04/15/nx-s1-5355895/doge-musk-nlrb-...
If the allegation is true, what would be the motivation of the higher-ups to keep this secret from US-CERT?
It appears to be a severe compromise, and the context suggests that much of the rest of the federal government is imminently vulnerable to the same tactics by the same threat actor.
Where the higher-ups reporting the security crisis through better channels?
Or were they trying to keep it quiet entirely, so might be complicit in something bad?
This appears to be DOGE employees simply doing their job.
You may not agree with what they’re doing in a political sense, but if you were tasked with the same problem you’d come up with a nearly identical solution.
For example: “tenant admin” is probably the special role that can bypass access control (not audits!) and see and read all data.
This sounds scary but I regularly request this right from large government departments and I get it granted to me.
Its use is justified when normal access requests would be too complex / fiddly and error prone. Generally, in a large environment, there is no other way to guarantee 100% coverage because as an outsider you don’t even know what permissions to ask for if you can’t see anything due to a lack of permissions!
Seriously: sit down for a second and think about how you would go about getting access to make a full copy of an organisation’s data for an audit if you fully expect both passive resistance and even active efforts to hide the very things you’re looking for.
"7. March 3rd - I received a call during which an ACIO stated instructions were given that we were not to adhere to SOP with the doge account creation in regards to creating records. He specifically was told that there were to be no logs or records made of the accounts created for DOGE employees."
Which part of doing an audit, or some other DOGE employee's job, requires logs or records not to be made of their accounts?
Another quote:
"They were to be given what are referred to as “tenant owner” level accounts, with essentially unrestricted permission to read, copy, and alter data. Note, these permissions are above even my CIO’s access level to our systems. Well above what level of access is required to pull metrics, efficiency reports, and any other details that would be needed to assess utilization or usage of systems in our agency. We have built in roles that auditors can use and have used extensively in the past but would not give the ability to make changes or access subsystems without approval. The suggestion that they use these accounts instead was not open to discussion."
Audits don't require being able to alter data.
Also, some of the data is mentioned as being sensitive. Although granting access to the data of another agency may make sense, I have trouble believing that direct access to data such as sensitive personal information of third parties would routinely be given to people from outside of the organization. Even within the organization the group of people given access to sensitive data should be as limited as possible.
In fact, should the auditor find there is a way for them to access sensitive data without it being logged, they will flag it immediately. That would be the case even under simple financial regulation.
There is absolutely the risk that the people you audit will lie to you or present you with false data. In practice that's not common, because they stand to at the very least lose their jobs. It could also be illegal. Not worth it.
All of the public complaining is by staff that don’t understand their new position in the pecking order.
There is a King in charge and he cares not for the wailing of the petty nobles.
https://www.npr.org/2025/04/15/nx-s1-5355896/doge-nlrb-elon-...
Besides, no one needs unmonitored write access for audit. Even less DOGE who does no audit and don't have knowledge how to do audit. Audits are supposed to he traceable.
Also, in some cloud systems full read access can give you direct or indirect access to service keys / API keys which then are write equivalent permissions anyway.
DOGE employees aren't simply doing their job. They are actively subverting the government to fatally wound it.
Clearly the (system) auditing infrastructure wasn't robust enough to still provide a lot of monitoring even in the service is being managed by someone else...
Also a several hundred line teardown of a 300line file is exactly what is wrong with some coders. Not having a CI/CL for every single short tool written once to do a job is called being productive...
Prove it. I want you to give examples of where you did something like this.
Why is anything of significance on github in the first place?
Edit: It's not. They just download python libraries to do "IP rotation" to circumvent rate limits.
On the actual complaint: (https://whistlebloweraid.org/wp-content/uploads/2025/04/2025...)
It seems that the data was stored in Azure which doesn't make it any better.
They downloaded "IP rotation" python libraries to circumvent rate limits.
Guessing those are the same accounts that got accessed by Russian IPs?
Genuinely wondering whether the US democracy is going to make it to December.
That isn't what "open source" means.
Let's start with this:
> Berulis said the new DOGE accounts had unrestricted permission to read, copy, and alter information contained in NLRB databases.
> Berulis said he discovered one of the DOGE accounts had downloaded three external code libraries from GitHub
What exactly does that mean? NLRB database accounts are GitHub accounts? (Surely not.) Or the same IP address accessed both, suggesting it was the same person? Define "account".
No coherent point being made here. This story needs to clearly separate the rhetoric about GitHub repositories from the NLRB access, and connect them together coherently.
The flow seems to be:
1. Some DOGE people obtained unbridled access to NLRB, with the ability to erase audit trails.
2. There is some sort of evidence that the same people downloaded tools from GitHub for distributed web scraping, suggesting intent to scrape massive amounts of data from somewhere (inferred to be the NLRB database).
There is no evidence cited in the article for the actual downloading of gigabytes of data; the "whistleblower" is quoted only as saying that DOGE required certain privileged accounts to be created and that the users of the accounts supposedly downloaded some web scraping software from GitHub.
At least mention some circumstantial evidence, like a suspicious increase in access activity, coming from distributed IP addresses in the Amazon cloud, following the download of those tools.
This:
> On February 6, someone posted a lengthy and detailed critique of Elez’s code on the GitHub “issues” page for async-ip-rotator, calling it “insecure, unscalable and a fundamental engineering failure.”
seems neither here nor there; why include that. It may be that the tools DOGE are using are not adequately safeguarding the data, but it seems like an extraneous point, and undigestable without specifics.
> Upon learning of your resignation, following reports that you were linked to an account advocating to “normalize Indian hatred” and for a “eugenic immigration policy,” I can’t help but address the staggering hypocrisy of these views within the context of the IT industry.
> This field, including your own career, is built on the labor, innovation, and expertise of Indian engineers and developers. To hold such hateful beliefs about a group that forms the backbone of this industry isn’t just reprehensible—it’s a complete contradiction of the reality you benefit from every day.
> My original critique of your code addressed technical issues and provided solutions, but after learning about your expressed views, it’s clear that poor coding isn’t the root problem here. Your mindset is incompatible with the fundamental values of IT: collaboration, respect, and global interconnectedness.
> Someone who advocates for hate cannot build systems meant to serve diverse users, nor can they lead or contribute meaningfully to teams that rely on trust and mutual respect. I strongly suggest you reflect on the harm your beliefs cause—not just to others, but to your credibility and future in this profession.
It doesn't invalidate the same author's critique above it at all (the critique itself manages to do that) but how it ended up mentioned in Krebs' article is puzzling. It harkens back to the days when journalists would quote-mine random Twitter users' tweets as if it meant something. "Twitter user @john89674651684685 said…" Give me a break.
[0] https://web.archive.org/web/20250423135719/https://github.co...
Firstly, anyone claiming that "the whole government is compromised" is being conspiratorial. Breaches of this nature are reportable to CISA (US-CERT), the DOJ, local law enforcement, and the FBI. The NLRB has its own cybersecurity incident response team, which includes legal counsel. If both the NLRB and US-CERT determined that this wasn’t a reportable incident then I trust their judgment.
Secondly, I’ve seen a lot of speculative commentary about the Russian IP allegedly logging into the DOGE account. A simple OSINT investigation reveals that this IP has had a negative reputation for over a year, specifically flagged for credential stuffing and scanning activity. Credential stuffing is a common tactic when credentials have been leaked or breached, often showing up on platforms like intelx.io, DeHashed, or BreachForums.
It's also worth noting: no serious nation-state actor would use an IP with such a known bad reputation. Doing so would risk burning any operational investment they’ve made. Nation-state actors almost always use clean infrastructure or proxy chains to conceal their activity.
The timeline the whistleblower presents spans two months, yet I find his interpretation of the activity speculative without hard evidence—especially considering he admits he does not possess the actual logs. That’s a huge red flag.
Thirdly, I tried to find the whistle blower’s official title, and it’s usually hidden in the media. In his official report he states that he is a Dev Sec Ops engineer. He also claims that he lost access to privileges – but the emails in the screen shot seemed to be a zero-trust/principle of least privileges hardening effort. That’s not suspicious to me.
Fourth, the screenshots the whistleblower provided of the Azure environment appeared extremely sparse. While I don’t know the exact size of the NLRB’s infrastructure, unless it's unusually small, I would expect to see more resources. From what I reviewed, the Azure dashboards he used had no filters applied, which raises the question—why are there no other subscriptions, VMs, load balancers, WAFs, etc., visible?
Regarding the DLP policy alerts, he could have easily shown the associated data. Interestingly, the alerts were labeled “test,” which is significant—but he chose not to address or explain that. Omitting that context makes the evidence less compelling. He also leaves out basic critical Indicators of Compromise (IOCs) like src_ip, src_port, dest_ip, dest_port, bytes, and duration. I’m not expecting him to extract mutex and environment variables but showing the basics would be convincing enough consider all they would have been accessible to him from the dashboards he screenshots in the document.
Finally, his claim that the NLRB doesn’t have a SIEM is demonstrably false. The NLRB shares a SIEM with the DOJ, which is operated by MindPoint Group under a SOCaaS contract.
Here’s my general take on the situation: The whistleblower had only been with the organization for six months and served as a mid-level DevSecOps engineer—not a security analyst, incident responder, or SOC analyst. After DOGE was announced, the NLRB began implementing Zero Trust principles and the Principle of Least Privilege. This is typical hardening. As a result, his old admin access which was over provisioned and no longer necessary for his role—was revoked. He panicked. Still having access to some Azure tools, he could have used a test or dev environment (referencing the sparse number of resources in the screenshot but he claimed it to be prod with no filter), toggled a few settings, took screenshot, and constructed a narrative around it. He escalated it to the CEO, who initially listened. However, the incident response team conducted an investigation and found nothing substantiating his claims. NLRB and US-CERT determined it to not be reportable, or which indicates that if it was a security event it was not an incident.
As for the Russian IP, it may be real—but it’s clearly tied to credential stuffing activity, not a sophisticated threat actor. If it genuinely accessed a DOGE account, that would indicate a breach on the DOGE side or weak password hygiene. But again—as mentioned earlier—he doesn’t have the logs to back this up, and his reasons for that are unconvincing. #Doubt.
The evidence supporting his claim is a screenshot of an Excel spreadsheet with several columns excluded. It appears to have been exported from the DeviceProcessEvents table within the advanced threat hunting schema. However, he failed to provide the threat hunting dashboard view, which would include critical context such as the process tree, MD5 hash, account SID, account domain, and process creation time. Given that he clearly has access to Microsoft Defender XDR or Defender for Endpoint, he has the capability to conduct a thorough investigation. Yet, he did not do so, nor did he include that information in his legal submission. As a result, I find his claims unconvincing.
As for the forked repo deletion - I have no clue. It seems like the repo was already well known. I'm not a dev so I'd defer to a dev's opinion here. The system owner could be function testing, fuzzing, performance testing, ect. Why didn’t he show the process tree, the system name, and netflow to prove that system running code was interacting with prod? – He clearly has access to Azure tools that would allow him to do that.
The (under oath) claims of extraction of data seem strange for the reasons you mentioned but so do the threats as well as the NLRB PR rep stating that DOGE was never there, I think there's more to be discovered that could clarify what happened.
Now, the govt also has to create rules for itself. So it creates the Privacy Act and layers of beurocratic checks and balances. These rules are to protect the people, not to derisk or protect the govt. After all, the govt has all the power.
So when capitalist businesses leaders are given the keys to govt, the normal ways of ethical alignment don't work. If you don't follow your own rules, who cares? They're your rules! I think what we're seeing is what happens if you apply traditional capitalist business practices to govt administration.
In some countries, this is done with outright bribery. Here, we do it with campaign contributions and lobbying and “we’ll create jobs in your district.”
Honestly, if you were around watching the news 30+ years ago, you would notice a stark difference in how news is covered then versus today. You can't really blame them, they are doing what they can to survive, but coverage today much more tabloid than news.
I would say the "fake but accurate," was the death knell, but it might have been sooner.
Also interesting to note that not only has Berulis' attorney lead multiple lawsuits against the Trump administration in the past, he was also an intern for both Chuck Schumer and Hillary Clinton. Now that obviously doesn't prove anything, but it could nonetheless be considered a strong indicator this all might be politically-motivated.