Rendered at 10:13:46 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
halestock 19 hours ago [-]
Pretty impressive to introduce 1400 CVEs in a project that's only ~7 months old.
GavCo 19 hours ago [-]
These are CVEs in the base image and in standard lib dependencies. For example, just scanned an unhardened image I built today:
Unhardened: docker.io/nanoco/nanoclaw:agent-alpha
71 packages, 344 unique CVEs, linux/arm64
PACKAGE VERSION TYP C H M L N TOT
-----------------------------------------------------------
expat 2.5.0 deb 0 4 18 1 2 25
curl 7.88.1 deb 4 4 6 0 7 21
hono 4.12.14 npm 0 1 18 2 0 21
libtiff 4.5.0 deb 0 2 1 1 15 20
perl 5.36.0 deb 5 6 3 0 3 17
pnpm 10.33.0 npm 0 8 7 0 0 15
glibc 2.36 deb 1 2 2 1 7 13
openjpeg 2.5.0 deb 0 0 3 1 9 13
cups 2.4.2 deb 0 2 8 0 1 11
glib2 2.74.6 deb 1 7 1 0 1 10
tar 1.34(+2) deb 1 1 7 0 1 10
llvm 15.0.6 deb 0 0 0 1 9 10
sqlite3 3.40.1 deb 1 2 3 0 3 9
nss 3.87.1 deb 1 0 3 0 4 8
avahi 0.8 deb 0 0 8 0 0 8
util-linux 2.38.1 deb 0 0 3 0 2 7
elf 0.188 deb 0 0 0 0 7 7
libssh2 1.10.0 deb 1 4 1 0 0 6
openldap 2.5.13 deb 0 1 0 0 5 6
chromium 151.0.7922.108 deb 0 5 0 0 0 5
-----------------------------------------------------------
UNIQUE CVEs 16 68 121 17 119 344
(+51 more packages, 102 findings)
C/H/M/L/N = critical/high/medium/low/negligible.
Counts are unique CVEs: binaries from one source package are
grouped (libcurl4 + libcurl3-gnutls + curl = curl), so a CVE
hitting three of them counts once, not three times.
m4rtink 2 hours ago [-]
So it has too many dependencies l, which are themselves also CVE magnets ?
Maybe they could depend on less items that are more secure ?
17 hours ago [-]
viccis 18 hours ago [-]
Are these real findings, or a situation in which fixes have been backported? At one place I worked, the corpsec guys were wildly incompetent and would try to bury me in "CVEs" in my systems that were nothing but "vulnerable" software versions with all of the "identified" vulnerabilities fixed by Debian backported patches.
hylaride 17 hours ago [-]
Most modern scanners know the backported versions for the major vendors (ubuntu/debian/rhel/alpine), but it was definitely an issue a decade ago.
lokar 17 hours ago [-]
That’s not security, it’s compliance.
17 hours ago [-]
devin 19 hours ago [-]
If the thing measuring whether there are CVEs is also the thing creating said CVEs, are we sure they are even CVEs? Deduped? Etc.
19 hours ago [-]
19 hours ago [-]
mohamedkoubaa 17 hours ago [-]
Hold my beer
random3 18 hours ago [-]
It’s like it’s made of CVEs. First 50-100 should be a good sign if it’s cleaner to start over.
prymitive 17 hours ago [-]
For those unfamiliar, “CVE” stands for “CV Enrichment”, common slang in Posture Engineering
Discovered a new legit CVE today during a meeting with some other engineers. I’m going to make sure the one who originally brought it up gets to put it on his résumé. The world needs more people like that.
482937632992 17 hours ago [-]
"'They' just don't want you to know how incredibly secure this vibeslopped clawshit is."
tptacek 17 hours ago [-]
If you're not a security person, the unspoken subtext here: the overwhelming majority of these "CVEs" do not matter to the project, and a very large number of them don't matter at all. They're pro-forma findings, like ReDOS in code paths that are rarely used, or, even more commonly, "prototype pollution" issues.
aliasxneo 18 hours ago [-]
I'm convinced you can tackle 5-10 "CVEs" a day, make a little dashboard, put some pretty graphs on it, and send it to your exec team and probably get accolades. Nevermind that the CVEs had nothing to do with your product.
lokar 17 hours ago [-]
My favorite urgent must fix CVE from compliance was a bug in the Linux PCMCIA driver on some EC2 VMs.
chrismorgan 16 hours ago [-]
I don’t say it’s the case there (it probably isn’t, you probably need hardware access or root), but sometimes those sorts of things do actually matter because there’s a way of causing them to be run anyway.
This is why it’s good to exclude things you don’t need. The less there is, the fewer places there are for problems to lurk.
lokar 15 hours ago [-]
Not against fixing it, but at the next normal update, it an emergency.
sriram_sun 17 hours ago [-]
Why is that even part of the image?
lokar 15 hours ago [-]
It was a stock/default kernel setup, before my time.
nathancahill 18 hours ago [-]
This is how Vanta et al. make millions.
paulryanrogers 16 hours ago [-]
I don't think Vanta itself even scans for CVEs? They scan for compliance which includes tooling that may look for CVEs, like Dependabot.
Scanning companies should be sophisticated enough to distinguish invalid CVEs and back ported fixes.
evanjrowley 18 hours ago [-]
Why is the Node ecosystem like this? Why do people continue to choose it for popular projects vs. anything else?
tptacek 17 hours ago [-]
Because it's written in Javascript and there are classes of highly-situational and not-especially-meaningful vulnerabilities that broadly impact the entire ecosystem ("prototype pollution" is the canonical example) that get counted individually in every occurrence in every NPM package.
It's not a real number, and, worse, it obscures the real figure of merit (/demerit). The Node ecosystem does have a real problem with its culture of sprawling dependencies. But it's not the first-order issue with this "1400" number.
itintheory 18 hours ago [-]
It's the opposite of NIH syndrome. Need to left pad a string? Just import a library from some rando on the internet!
sophacles 17 hours ago [-]
When node was new there were many browsers (different rendering and js engines, not just chrome reskins) and standards and "proto standards" were moving very fast. Different browsers and versions would have very different support for CSS directives, tag behavior, etc. This was a real pain in the ass to make a site consistent across different browsers - every other line of code would require a full switch statement based on browser and version it seemed, and all of these things would need updating every time some browser had an update.
The answer to this was something called polyfills, a library that did something as simple as element.center() with just 500 lines of code to make it consistent on all browsers, and all the places you want to center that element are updated by the polyfill authors and your code doesn't need to be touched.
All the sites that weren't using good polyfills broke (or at least looked terrible) for days every time there was a new $browser update.
Since thats the javascript environment node was born into, the style was carried over by inertia and habit, for better and worse.
ljm 17 hours ago [-]
Because, like it or not, it does Write Once, Use Anywhere better than Java ever did.
It is pretty much the lowest common denominator for code.
shimman 17 hours ago [-]
Because it's written by a type of dev that thinks everything should use JS because that's what they used for 6 months while working at a FAANG company before getting laid off and since they worked at a FAANG company clearly they know better than you so if you attack them for using an extremely poor language you just aren't a seriously person so please shut up while I go fishing for some VC funds.
It's like asking why Claude Code is written with react, because the devs getting paid >$500k a year clearly know better d'uh.
eviks 17 hours ago [-]
Primarily because of the original sin of JS being an awfully designed language.
Partially because it's the most popular.
iandanforth 19 hours ago [-]
I don't understand the 'custom patch' strategy over 'fix the app with a major version change' strategy.
pokstad 11 hours ago [-]
I was intrigued by nano claws more “secure” marketing, but I couldn’t believe how loose and vibe coded the installation and set up process was. My god it’s full of prompts.
raver1975 17 hours ago [-]
That's what happens when you vibe code.
KaiserPro 19 hours ago [-]
so s/bookworm/trixie/g didn't work then?
Yes, this is mostly a joke, I am able to understand the difference between base distros.
iririririr 18 hours ago [-]
a spot on "joke". deb12 is gone for a month now.
eviks 17 hours ago [-]
What is NanoClaw? Glad you asked:
> NanoClaw is a secure, lightweight alternative to OpenClaw.
overgard 17 hours ago [-]
I'm pretty skeptical you can call any claw-like thing secure unless you solve prompt injection.
stavros 17 hours ago [-]
Security isn't binary. There's nothing that's "secure" unless you define a threat model first.
TZubiri 17 hours ago [-]
Sounds like the kind of stuff devs that introduce security vulnerbilities say.
r14c 12 hours ago [-]
if you can't define a threat model, or know why you need one, you will write vulnerable code.
stavros 16 hours ago [-]
Unrelatedly, I wonder why I don't see the downvote arrows on comments any more.
Hamuko 17 hours ago [-]
If the lightweight alternative has 1400 CVEs, how many does OpenClaw have?
sajithdilshan 17 hours ago [-]
I wonder how many new CVEs were introduced while patching these
bryan0 17 hours ago [-]
Why hasn't looking at EPSS (Exploit Prediction Scoring System) become a more standard approach than just raw CVEs?
Surac 17 hours ago [-]
let me guess. they wrote a promt that told claude do undo all bugs?
Maybe they could depend on less items that are more secure ?
https://en.wikipedia.org/wiki/Common_Vulnerabilities_and_Exp...
This is why it’s good to exclude things you don’t need. The less there is, the fewer places there are for problems to lurk.
Scanning companies should be sophisticated enough to distinguish invalid CVEs and back ported fixes.
It's not a real number, and, worse, it obscures the real figure of merit (/demerit). The Node ecosystem does have a real problem with its culture of sprawling dependencies. But it's not the first-order issue with this "1400" number.
The answer to this was something called polyfills, a library that did something as simple as element.center() with just 500 lines of code to make it consistent on all browsers, and all the places you want to center that element are updated by the polyfill authors and your code doesn't need to be touched.
All the sites that weren't using good polyfills broke (or at least looked terrible) for days every time there was a new $browser update.
Since thats the javascript environment node was born into, the style was carried over by inertia and habit, for better and worse.
It is pretty much the lowest common denominator for code.
It's like asking why Claude Code is written with react, because the devs getting paid >$500k a year clearly know better d'uh.
Yes, this is mostly a joke, I am able to understand the difference between base distros.
> NanoClaw is a secure, lightweight alternative to OpenClaw.