#web

### Summary Sanitization of the content of uploaded SVG files was not covering all possible XSS scenarios. ### Details DNN validates the contents of SVG's to ensure they are valid and do not contain any malicious code. These checks were introduced as part of `CVE-2025-48378`. However, the checks to ensure there are no script elements within the SVG files are not comprehensive and may allow some malicious SVG files to be uploaded. As this vulnerability allows for the execution of arbitrary JavaScript code within the context of the user's browser, it can lead to a range of attacks, including data exfiltration, session hijacking, and defacement of the web application to name a few.

### Summary The out-of-box experience for HTML editing allows unauthenticated users to upload files. This opens a potential vector to other security issues and is not needed on most implementations. ### Details The new out-of-box experience blocks that endpoint to unauthenticated users. If there is a real need for the implementation to allow unauthenticated uploads, then the web.config can be edited by the implementer to remove that block and open the endpoint to the public.

The second major cloud outage in less than two weeks, Azure’s downtime highlights the “brittleness” of a digital ecosystem that depends on a few companies never making mistakes.

### Summary SQL Injection vulnerability in TypeORM before 0.3.26 via crafted request to repository.save or repository.update due to the sqlstring call using stringifyObjects default to false. ### Details Vulnerable Code: ```js const { username, city, name} = req.body; const updateData = { username, city, name, id:userId }; // Developer aims to only allow above three fields to be updated const result = await userRepo.save(updateData); ``` Intended Payload (non-malicious): ` username=myusername&city=Riga&name=Javad ` _OR_ `{username:\"myusername\",phone:12345,name:\"Javad\"} ` SQL query produced: ```sql UPDATE `user` SET `username` = 'myusername', `city` = 'Riga', `name` = 'Javad' WHERE `id` IN (1); ``` Malicious Payload: `username=myusername&city[name]=Riga&city[role]=admin ` _OR_ `{username:\"myusername\",city:{name:\"Javad\",role:\"admin\"}} ` SQL query produced with Injected Column: ```sql UPDATE `user` SET `username` = 'myusername...

Hackers exploit critical XWiki flaw CVE-2025-24893 to hijack corporate servers for cryptomining, with active attacks confirmed by VulnCheck researchers.

Infamous botnets like Mirai are exploiting Web-exposed assets such as PHP servers, IoT devices, and cloud gateways to gain control over systems and build strength.

Cybersecurity researchers are calling attention to a spike in automated attacks targeting PHP servers, IoT devices, and cloud gateways by various botnets such as Mirai, Gafgyt, and Mozi. "These automated campaigns exploit known CVE vulnerabilities and cloud misconfigurations to gain control over exposed systems and expand botnet networks," the Qualys Threat Research Unit (TRU) said in a report

Jenkins SAML Plugin 4.583.vc68232f7018a_ and earlier does not implement a replay cache. This allows attackers able to obtain information about the SAML authentication flow between a user’s web browser and Jenkins to replay those requests, authenticating to Jenkins as that user. SAML Plugin 4.583.585.v22ccc1139f55 implements a replay cache that rejects replayed requests.

Cybersecurity researchers have flagged a new security issue in agentic web browsers like OpenAI ChatGPT Atlas that exposes underlying artificial intelligence (AI) models to context poisoning attacks. In the attack devised by AI security company SPLX, a bad actor can set up websites that serve different content to browsers and AI crawlers run by ChatGPT and Perplexity. The technique has been

No, Gmail wasn’t hacked. But a flood of old stolen credentials on the dark web sparked headlines suggesting otherwise. Here’s what really happened.