SShortSingh.
Back to feed

Researchers Disclose Universal XSS Flaw in uBlock Origin via Website-Driven Element Picker

0
·4 views

A security write-up describes a universal cross-site scripting (XSS) vulnerability in the uBlock Origin content blocker. The flaw is reportedly triggered through the extension's element picker when driven by a website. The disclosure was shared on Hacker News, where it gained modest attention. The full technical details are available in the linked post on iter.ca. The article text provided is limited to the link and metadata, so further specifics about affected versions or fixes could not be confirmed.

Read the full story at Hacker News

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Learning Web Table Accessibility: Captions, Structure, and Scope Attributes

A developer documents their study of advanced HTML table features using MDN's table accessibility guide. The notes cover the caption element and the thead, tbody, and tfoot sections, which add structure and make styling easier. They also review scope attributes (col, row, colgroup, rowgroup) and the id and headers attributes for linking cells. The author notes that many of these features make little visible difference but improve support for screen readers. They plan to work through MDN's Tables Assessment lesson the following day.

0
ProgrammingDEV Community ·

AI search growth puts pressure on the web indexes that answer engines depend on

AI-driven search traffic has grown sharply since 2024 and now makes up a small but rising share of website visits, while Google's share of search has slipped below 90 percent. Every AI answer engine draws on an index, and a few companies control most of those indexes, so pages missing from them are effectively invisible. Automated traffic already exceeds half of all web traffic, according to a 2026 bot report. The independent search engine cl0q.com described how a surge of requests from a single scraper fleet caused thousands of server errors in mid-November, prompting it to introduce rate limits and disk caching. The company argues that smaller, independent indexes could reduce the web's dependence on any single company's crawl.

0
ProgrammingDEV Community ·

LangChain Raises $300M at $3B Valuation as AI Startups Shut Down

On November 13, LangChain raised $300 million at a $3 billion valuation, led by Alphabet's CapitalG, more than doubling its worth. The company's LangSmith platform for agent observability and management is reportedly used by 7,000 organizations. In the same period, several AI product companies have closed, including Character.ai, which filed for bankruptcy in October, and Ollie, which shut down in November after five months. The article argues investors are backing infrastructure such as tracing, evaluation, and audit tools over consumer-facing AI products. It cites S&P Global Market Intelligence estimating that about 42% of AI initiatives are abandoned by the companies that start them.

0
ProgrammingDEV Community ·

TinyGIF converts and shrinks GIFs in the browser, splitting a large FFmpeg binary to fit Cloudflare

Danny, the builder of TinyGIF, describes a free GIF tool that processes files entirely on the visitor's device so nothing is uploaded to a server. The core FFmpeg WebAssembly binary is about 30.7 MiB, which exceeds Cloudflare Workers' 25 MiB limit for a single static asset. To work around this, the file is split into smaller parts at build time, downloaded in parallel by the browser, and reassembled before compilation. Existing GIFs are compressed with gifsicle compiled to WebAssembly, while video-to-GIF conversion uses FFmpeg in a Web Worker. The approach trades a one-time converter download and limited memory on phones for privacy and zero per-file server costs.