From my experience I have not experienced customer feed / customer thinking spiralling towards drugs and gambling tips.
Then also, companies are guided by some mission / values which might prevent you from going into that direction, unless you are working for an online gambling company
I have full control over the inputs for hugo and the output is pure static HTML. It's better if there are no CVEs but I really don't sweat these. My Hugo template runs 100% pure CSS and no JS.
This seems like a really good example of Base CVSS scores not telling us much on their own.
For the Wordpress RCE (nominally CVSS 9.2), it looks like many standard deployments of WordPress would be affected, barring extra mitigations. But in the case of these Hugo ones (9.3), it looks like very specific circumstances (anti-mitigations, if you will) are needed. E.g. running arbitrary builds of untrusted user content without a sandbox; running it in a GitHub workflow against PRs from untrusted contributors, etc.
> These metrics enable the analyst to customize the CVSS score depending on the importance of the affected IT asset to a user’s organization, measured in terms of complementary/alternative security controls in place, Confidentiality, Integrity, and Availability. The metrics are the modified equivalent of base metrics and are assigned metric values based on the component placement in organization infrastructure.
You need to reason about both probability and impact, not one or the other. In your example, it just means that the probability is very low, perhaps even down to 0 in your specific case. But even at a low probability, the impact of the vulnerability remains very high so it helps you reason about it.
For example, you might react differently to these scores:
- <8/10: check that your systems are indeed secure
- 8.6/10: check that your systems are indeed secure and tell your junior analyst to train on creating a custom monitoring rule for that attack and follow-up with you
- 9.8/10: double-check that your systems are indeed secure, ensure that if you had a hole another security layer would have caught it (if not, that's a problem!), set up a honeypot to get some info on the assholes that have repeatedly attacked you lately and will undoubtedly try to 0-day you in the next few hours, etc.
Ironically, this 9 years old comment on the official documentation page of one of the affected functions perfectly describes both the nature and remediation of this major security flaw:
> Paul Ryan 9 years ago
> Note that locate_template() does not prevent directory traversal attacks, so if you’re passing a user-provided template name to the function, be sure to verify that it’s from one of the three appropriate locations (active theme directory, parent theme directory, or /wp-includes/theme-compat/ directory).
Ahahah I remember to have patched themes for clients by hand, years ago. A different time, where a core team would for whatever reason leave security holes around to be sure you need three frameworks around their pile of dung code.
> This is a somewhat situational Wordpress RCE that impacts only a couple themes.
reply
That is dangerously incorrect, a whole lot of themes are vulnerable. The main pre-condition, "presence of a top-level directory named 'page-xxx' like 'page-templates' in the theme's directory" is actually an official recommendation in the WordPress documentation.
> As discussed in Organizing Theme Files, WordPress can recognize page templates stored in the theme’s root folder or in a first-level subdirectory of the theme folder. *The page-templates/ folder is a common convention* for organizing global page templates, but it is not required. Page templates can also be stored in other first-level subdirectories, such as templates/ or page_templates/.
I'm not sure I understand. That's the main pre-condition... to include an arbitrary PHP that is already in the server.
On a fresh WP install, a random user can't upload PHP files. Normally you don't even need to allow random users to register an account since avatars on comments come from gravatar anyway.
> On a fresh WP install, a random user can't upload PHP files.
Indeed, but you don't need to upload anything as long as there's already a PHP file that allows you to execute arbitrary commands somewhere on the server, right?
Well, as it turns out, the default PHP Docker image has had such a file readily available until version 8.5 =)
Jesus. I don't think WP is nice to use, but people are saying WP is crap because someone else made pearcmd.php that allows you to run arbitrary commands, and then a third someone else included that in the default PHP docker image. In this case all PHP CMS's are a directory traversal bug away from this CVE.
Indeed. Then again, it's on WordPress not to have directory traversal flaws in their core functions...
- especially the functions that are explicitely exposed to be used on front-facing interfaces (ie: templating functions).
- and especially when the security flaw in question was not only raised 9 years ago but described in details on the official documentation page of the affected function
Yeah not sure why you’re being downvoted when the built in themes are vulnerable as is the default docker image pre php8.5 which a lot of people use as a base and I bet a lot of hosting providers use as well behind the scenes.
On some benchmarks models like Qwen 3.8 Max which cost < $6/m out cost more than Astra 6 to run at $50/m out. That’s a huge price gap and yet Astra would be cheaper if your work looks like the benchmark.
We notice with frontier models like Astra and Fable that one might use a lot less tokens than the other to complete the task thereby being the better deal in spite of the far higher token cost.
Astra on xhigh has a cost per task of $2.31 with an intelligence index of 53. Qwen3.8 Max has a cost per task of $5.41 with an intelligence index of 45. Pricing for GPT-6 Astra (xhigh) is $10.00 per 1M input tokens and $50.00 per 1M output tokens. Pricing for Qwen3.8 Max (0902) is $2.00 per 1M input tokens and $6.00 per 1M output tokens.
Obviously this is just one measure of all of this (and Qwen 3.8 Omni Flash isn't yet available), but I think this illustrates the point well. These relative task costs are pretty consistent across different analysts. Cost per token is arguably a useless measure at this point in most circumstances.
Did you open the repo? There's a subdirectory with two dozens of files, around 300Kb of text.
Storing/visualizing small text changes over time as revisions is exactly what Git is excellent at, how else would you keep track of updates to the prompts?
Indeed. It's been very funny to observe introvert geeks progressively rediscover human communication over the past 3 years.
For decades, they were allowed -expected even- to be weird and communicate badly with their peers, let alone third parties.
Then, a whole zoo of processes, gamifications and other shenanogans were invented just to help them show normies what was up with their work (remember the planning poker game?)...
Now, at last, those people are actually ewpected to be able to explain their constraints, document the work, be accountable for the results and generally exchange contructively... Just, well, with AIs not with humans.
I have not seen this kind of Claude way of speaking in professional communications aside from PR stuff or etc. to the contrary, its often quite casual but still polite
if you're asserting Claude-like speech is how normies speak?
reply