All Topics
All Topics
Technology
Technology
Design
Design
Programming
Programming
Science
Science
News
News
Gaming
Gaming
Entertainment
Entertainment
Business
Business
Finance
Finance
Sports
Sports
Health
Health
Food
Food
Travel
Travel
Art
Art
Music
Music
Books
Books
Education
Education
Politics
Politics
Personal
Personal
Bluesky
Twitter
No algorithm. No AI slop. No ads. Just RSS. Pro-human. Indie writers. Real journalism. Open web. Chronological. Hand toasted.

The internet, freshly baked.

5,341 publishers in the oven.
32,124 feeds, checked for AI clickbait slop.

Try itMediumBBCPrototyprWikipedia
Why FeedBagel

Pro-human. Not anti-AI.

The web used to be full of weird, brilliant, hand-written stuff. Then the feed got hijacked by ranking models, engagement bait, and highly-processed AI slop with no real substance. No nutrition, no calories, just keywords tuned for SEO.

We're trying to pick out the parts of that web still worth reading. So far: 5,341 publishers, 32,124 feeds, naively scored against clickbait and content-padding. AI helps in the kitchen. Slop gets ranked down, whoever wrote it. Better sources, better answers, whether the reader is human or the agent working for them.

What we believe
Human input
Bloggers, journalists, scientists, indie zines. Every feed points back to someone who chose what to publish.
RSS, chronological
RSS feeds, newest first. Any ranking is for quality, never engagement farming.
No slop, by anyone
AI can write. Humans can write. Padded, soulless content is a soggy bagel: it ranks low. Anti-AI-slop detection coming.
Built for agents too
API, MCP, webhooks. We want AI agents pulling from hand-picked sources, not the slop. Better answers come from better reading lists.
The directory

Browse by topic

A taste of the directory. Pick a topic to see the full list.

What we do

Find your daily bagel.

FeedBagel is a fresh search engine for RSS. Browse curated feeds, paste any site to fish out its hidden RSS, or type a topic and we toast a reading list for you. No algorithm, no ads, just the open web.

Multi-source synthesis

Baker's Take

When several outlets cover one story, we synthesize the coverage into a single piece.

Technology

Snap's SPECS Glasses Arrive After a Decade of AR Development, Blending AI and Mixed Reality

Snap Inc. unveiled its long-anticipated augmented reality glasses, called SPECS, at the Augmented World Expo 2026 in Long Beach, California, marking the culmination of over a decade of work in the field. The lightweight device is designed to merge mixed-reality display technology with AI assistant features, aiming to make AR a practical part of daily life rather than a niche gadget. According to Hacker News, the glasses are intended to help users "create, connect, learn, and accomplish tasks while staying present in the real world," continuing Snap's vision of using technology to deepen human connection. Dezeen reported that the product represents Snap's effort to blend the immersive qualities of mixed-reality headsets with the practical, everyday form factor of AI glasses. "The product represents Snap's effort to blend the immersive qualities of mixed-reality headsets with the practical, everyday form factor of AI glasses." This positioning sets SPECS apart from bulkier competitors, as Snap aims to make AR a seamless accessory rather than a cumbersome headset. The glasses allow users to enjoy entertainment, work on the go, and access AI assistance through an integrated display, according to Dezeen. Hacker News noted that SPECS are designed to blend AI assistance, work tools, entertainment, and shared experiences into everyday life. The announcement at the Augmented World Expo signals Snap's confidence that the technology is finally ready for a broader audience, after years of iterative development and earlier prototypes like the Spectacles camera glasses. "The glasses are designed to help users create, connect, learn, and accomplish tasks while staying present in the real world." With SPECS, Snap is betting that consumers are ready to wear AR daily, a bet that will test whether the company can translate its social media dominance into hardware success. The product's lightweight design and focus on practical use cases could help it avoid the pitfalls that have plagued other AR efforts, but competition from Meta and Apple remains fierce in a market still searching for a breakout hit.

2 sources · 19 Jun

augmented realitybusinesstechnology

Microsoft unveils Project Solara, an Android-based OS for AI agent hardware

Microsoft announced Project Solara at Build 2026, a new operating system built on Android rather than Windows, designed specifically for gadgets that run AI agents. The Verge reported that the company demonstrated two concept devices: a desk hub similar to an Echo Show with facial recognition for accessing AI agents, and a wearable work badge equipped with a camera and fingerprint sensor. "Microsoft positions Solara as a platform built from the ground up for agent-driven experiences." GeekWire noted that Microsoft has been secretly developing Solara as a platform for devices that run AI agents instead of traditional apps. The project has produced two working hardware concepts that are headed to pilots with major businesses, signaling a significant shift in Microsoft's strategy. "The initiative represents a significant shift in Microsoft's strategy, moving beyond Windows to embrace Android for AI-first devices." This move away from Windows for AI agent hardware marks a notable departure for Microsoft, which has long relied on its flagship OS. By adopting Android, the company appears to be prioritizing compatibility and developer access over its own ecosystem for these new agent-driven experiences. The pilots with major businesses will test whether Solara can deliver on its promise of transforming how users interact with computing through agent-based interfaces, according to GeekWire.

2 sources · 19 Jun

businesstechnology

Uber reins in AI coding tool spending with $1,500 monthly employee cap after budget blowout

Uber has imposed a $1,500 monthly spending cap per employee on agentic AI coding tools, reversing its earlier encouragement for staff to use artificial intelligence as much as possible after the company blew through its AI budget in just four months, according to TechCrunch. The limits apply specifically to tools like Anthropic's Claude Code and Cursor, and spending on one tool does not affect the budget for another, Bloomberg reported. The caps are tracked through an internal dashboard, TechCrunch noted. "This marks a reversal from the company's earlier encouragement for staff to use AI as much as possible, highlighting the growing cost challenges companies face with widespread AI adoption." Uber's move underscores a broader tension in the tech industry: companies want to push AI adoption to boost productivity, but the costs of running and licensing these tools can quickly spiral out of control. The $1,500 per employee per month cap is a concrete attempt to balance innovation with fiscal discipline. Bloomberg reported that the caps apply only to agentic coding software, meaning tools that can autonomously write and execute code. This distinction suggests Uber is targeting the most expensive and resource-intensive AI tools while leaving other AI spending potentially uncapped. As AI tools become more powerful and more widely used, other companies may face similar budget pressures. Uber's experience serves as a cautionary tale: even a tech giant with deep pockets can be caught off guard by the costs of widespread AI adoption, forcing a rapid shift from encouragement to control.

2 sources · 19 Jun

businesstechnology

Gaming

U.S. Gaming Population Hits 212 Million as Average Player Age Reaches 37

A new report from the Entertainment Software Association (ESA), conducted with YouGov, reveals that 212.3 million Americans, or two-thirds of the population, now play video games for at least one hour per week. According to Variety, this represents a 3% increase of 7.2 million people from the previous year, signaling continued growth in the industry's reach. The average age of gamers has risen to 37, according to Variety, while polygon.com reports the average age as 36. The report, which covers players ages 5 to 90, found that 72% of players are 18 or older, underscoring the broadening demographic of video game enthusiasts in the U.S. The industry generated $57 billion in revenue in 2024, polygon.com noted. "90% of players say gaming helps them relax and unwind." This finding, reported by polygon.com, highlights the role of gaming as a stress-relief tool for the vast majority of players. The report also found that 74% of gamers play with others, and 61% play daily, suggesting that gaming is a deeply social and routine activity for many Americans. "97% of Americans see gaming as having benefits including mental stimulation, problem-solving, and stress relief." According to polygon.com, the near-universal recognition of gaming's benefits points to a shift in public perception, with the activity now seen as a mainstream, cross-generational pastime rather than a niche hobby. The ESA's data, as covered by both outlets, reinforces that video games have become a central part of American entertainment culture.

2 sources · 19 Jun

entertainmentgaming
Just Indexed

Recently toasted

Sites whose feeds we just discovered and toasted.

More fresh takes

Entertainment

Amazon MGM Drops Nearly Finished Sam Altman Biopic After Inking $50 Billion OpenAI Deal

Amazon MGM Studios has dropped Luca Guadagnino's nearly finished film "Artificial," a biopic about OpenAI CEO Sam Altman starring Andrew Garfield, just months after Amazon struck a major strategic partnership with the AI company. According to Variety, the decision comes after Amazon invested $50 billion in OpenAI and expanded its use of Amazon Web Services in February, creating a clear conflict of interest between the studio's creative project and its corporate relationship. "The film will be better served if it were released by a different studio." Rolling Stone reported that Amazon's statement acknowledged the tension, noting the film would be better off elsewhere. The Hollywood Reporter added that Prime Video and Amazon MGM Studios chief Mike Hopkins personally informed Guadagnino and the producing team of the decision, leaving the project's future uncertain as it will now be shopped to other studios. The film centers on the dramatic five-day period in 2023 when Altman was briefly ousted as CEO of OpenAI and then reinstated, as reported by The Verge. Engadget confirmed the film's focus on that leadership crisis, with Andrew Garfield portraying Altman, alongside Monica Barbaro, Ike Barinholtz, and Yura Borisov in key roles. The move raises broader questions about how corporate partnerships can influence creative content. As Rolling Stone explored, the decision highlights the tension between Amazon's business relationship with OpenAI and its prior commitment to a film that critically examines the company's founder. The project now seeks a new home, with Variety noting it will be shopped to other studios in hopes of finding a distributor willing to release the nearly completed film.

5 sources · 19 Jun

businessentertainmentfilm industry

Warner Bros. Film Chiefs to Lead Inaugural TIFF Market Summit

The Toronto International Film Festival has announced that Warner Bros. Motion Picture Group co-chairs and CEOs Pam Abdy and Mike De Luca will headline the Speakers' Summit at TIFF's first-ever market, according to Variety. The pair will participate in a conversation with TIFF CEO Cameron Bailey on Sept. 11 as part of the Market's Dialogues program, with Variety serving as the media partner. "They will discuss the evolving role of studios in the global marketplace and developing award-winning content." The discussion will also cover the opportunities and challenges shaping the entertainment landscape and the future of film, as reported by Deadline. This marks a significant moment for TIFF, which is launching its inaugural market this year, aiming to position itself as a key hub for industry deal-making alongside its renowned film programming. Abdy and De Luca have been at the helm of Warner Bros.' film operations since 2022, overseeing a slate that includes major franchises and award contenders. Their appearance at the TIFF Market underscores the festival's growing emphasis on industry networking and commerce, a shift that organizers hope will attract top-tier talent and executives from across the global film business.

2 sources · 19 Jun

businessentertainment

Nick and Charlie Face Their Future in Final Heartstopper Trailer

Netflix has released the official trailer for "Heartstopper Forever," the feature-length finale of the beloved teen romance series, set to premiere on July 17. The trailer offers fans a first look at the emotional conclusion of Nick Nelson and Charlie Spring's love story, as the couple confronts the challenges of growing up and moving on. "The film puts Nick and Charlie's relationship to the ultimate test as their love story concludes on screen." The trailer, released alongside a new key art poster with the tagline "Love Like There's No Ending," teases the bittersweet theme of young love ending, either through breakup or growing up, as Nick prepares for university and Charlie finds independence at school, according to The A.V. Club and Deadline. The finale will wrap up the series after its three-season run, with creator Alice Oseman writing the screenplay and Wash Westmoreland directing, Variety reported. "The trailer offers fans a preview of Nick and Charlie's farewell." Stars Kit Connor and Joe Locke are returning as executive producers for the finale, which Netflix announced earlier this month alongside a poster, Bleeding Cool noted. The trailer hints that the ending may not be as final as it seems, TVLine reported, leaving room for speculation among fans of the series based on Alice Oseman's graphic novels.

6 sources · 19 Jun

entertainment

Politics

Anthropic Pulls Two AI Models Worldwide After U.S. Export Order Targets Foreign Nationals

Anthropic has disabled global access to its Fable 5 and Mythos 5 AI models after the U.S. government issued an export control directive under national security authorities, ordering the company to suspend use by any foreign national, including its own employees. The directive, received at 5:21pm ET, forced Anthropic to take both models offline for all customers to ensure compliance, according to anthropic.com. The move marks the first time the U.S. has banned a company's own employees from accessing its technology based on nationality, as reported by Hacker News. "This abrupt order forces Anthropic to disable these models for all customers to ensure compliance, while other models remain unaffected." The directive, issued by the Trump administration, suspends access for any foreign national both inside and outside the United States, Bloomberg reported. Anthropic released a limited version of Fable 5 widely this week, while the more advanced Mythos 5 had tightly restricted access due to cybersecurity concerns, according to the San Diego Union-Tribune. The suspension came just three days after Fable 5's public launch, as noted by Le Monde. The action was triggered by a claimed "jailbreak" that reportedly exposed minor vulnerabilities, according to undercodetesting.com. Anthropic complied but argues the directive is based on a misunderstanding, stating the identified jailbreaking technique revealed only minor vulnerabilities, Ground News reported. Anthropic has publicly stated it disagrees with the government's handling of the matter, the San Diego Union-Tribune added. "This represents a major escalation in treating frontier AI systems as sensitive defense assets subject to real-time kill switches, affecting even the company's own employees and foreign nationals worldwide." While the practical impact is limited due to the models' recent release and limited availability, the move is described as unprecedented, Hacker News reported. The directive lacks specific details about the national security concerns behind it, according to anthropic.com, and the long-term ripple effects of this policy remain significant, as Hacker News highlighted.

7 sources · 19 Jun

ai regulationexport controlspolitics

Senators Press FCC to Halt Paramount-WBD Deal Over Foreign Ownership Fears

Three Democratic U.S. senators are calling on the Federal Communications Commission to block the proposed merger between Paramount and Warner Bros. Discovery from closing until a national security review of foreign investment in the deal is completed. Senators Cory Booker, Elizabeth Warren, and Adam Schiff sent a letter urging FCC Chairman Brendan Carr to conduct a thorough review before allowing the transaction to proceed, according to Deadline. "49.5% of equity in the parent company of CBS, CNN, and 28 broadcast stations could end up controlled by three foreign governments." The senators pointed to Paramount's April 24 petition seeking FCC approval for significant foreign investment, and requested that the FCC not act on the merger until the Committee on Foreign Investment in the United States (CFIUS) finishes its review, Variety reported. The $111 billion deal between Paramount Skydance and Warner Bros. Discovery has raised questions about potential foreign influence over major U.S. media assets. Booker, Warren, and Schiff expressed concern that the merger could allow foreign governments to gain control over a substantial portion of American broadcast infrastructure, including CBS, CNN, and 28 broadcast stations. The lawmakers are pressing the FCC to prioritize national security considerations before granting any approval, according to both outlets. This intervention adds a significant regulatory hurdle to an already complex merger, as the senators seek to ensure that foreign ownership does not compromise U.S. media independence. The FCC has not yet publicly responded to the senators' request, but the letter signals growing bipartisan scrutiny of foreign investment in American media companies.

2 sources · 19 Jun

businesspolitics

Sean Penn to Helm Untitled Capitol Riot Drama, Bradley Cooper in Early Talks to Play Officer

Sean Penn is set to direct an untitled film centered on a police officer whose life leads to his presence during the January 6, 2021, Capitol riot, with Bradley Cooper in early talks to star, according to multiple reports. The project, acquired by Warner Bros. in a negative pickup, reunites Penn with the studio behind his upcoming film "One Battle After Another." "an unexpected story about friendship" Variety described the film as only tangentially related to January 6, a framing that appears designed to minimize potential controversy around the subject matter. The A.V. Club noted that no deal is in place yet for Cooper, though Penn has written the screenplay and will direct. Rolling Stone reported that given Penn's political leanings, the film will likely be critical of former President Trump. The outlet also flagged potential production hurdles, though it did not specify what those might be. The project follows Penn's 2023 documentary on Ukrainian President Volodymyr Zelensky and his recent Oscar win for "One Battle After Another," according to The A.V. Club. "a police officer whose early life leads to his involvement in the January 6th Capitol riots" Deadline, which first broke the news, reported that the film is described as a biopic about an officer who defended the Capitol during the insurrection. The project marks Penn's return to Warner Bros., which also distributed his 2023 film "One Battle After Another." As of now, no official title or production timeline has been announced.

4 sources · 19 Jun

entertainmentpolitics

News

Louis C.K. Returns to Netflix With New Stand-Up Special 'Ridiculous'

Louis C.K. is set to debut his latest stand-up comedy special, 'Ridiculous,' on Netflix on June 30, marking his first major distribution deal since his career was upended by sexual misconduct allegations. The special features material he has toured for about a year, including jokes about putting his elderly father in a retirement home, according to Variety. "Netflix is bringing Louis C.K. back with a new stand-up special called 'Ridiculous' later this month, nine years after cutting ties with him following his admission of sexual misconduct." The A.V. Club reported that this move signals Netflix's apparent lack of concern about the controversy, drawing a parallel to the platform's continued work with comedian Tony Hinchcliffe. The outlet suggested that Netflix has adopted a more permissive stance on controversial comedians in recent years. Variety noted that the special marks Louis C.K.'s first stand-up special released by a major distributor since his career controversies. The comedian had previously been dropped by Netflix in 2018 after admitting to sexual misconduct, which had led to widespread backlash and a temporary halt to his mainstream career. The decision to bring Louis C.K. back onto the platform reflects a broader shift in Netflix's approach to programming, as the company appears increasingly willing to work with figures who have faced public scrutiny. The A.V. Club highlighted this trend, noting that the streaming giant has shown little hesitation in re-engaging with controversial talent. 'Ridiculous' will be available for streaming on Netflix starting June 30, giving subscribers access to Louis C.K.'s latest comedic material after a long absence from major distribution channels.

2 sources · 19 Jun

entertainmentnews

BBC Drops Ashley Cain Series After Guardian Reveals Abusive Tweets About Women

The BBC has canceled the second season of "Ashley Cain: Into The Danger Zone" after The Guardian uncovered historical misogynistic tweets from the presenter. Cain, a former professional footballer turned reality TV star, was being cultivated by the broadcaster to appeal to young male viewers, but the corporation confirmed it has "no plans" to air the series produced by True North, according to Deadline. "The BBC stated the messages were 'completely unacceptable' and confirmed it has 'no plans' to air the series." The Guardian reported that Cain, who hosts the BBC Three documentary series and travels to dangerous locations to interview young men on society's fringes, made abusive remarks about women in posts from 2015. The outlet documented that he allegedly referred to women as "slags," "sluts," "psychos," and "bitches." The BBC initially withheld commitment on the show after the investigation, but later moved to cancel it entirely. Cain, an MTV personality as well as a former professional footballer, had been praised by the BBC for his appeal to young male audiences. The Guardian's investigation revealed the extent of the remarks, which the BBC deemed incompatible with its values. The cancellation marks a swift reversal for a series that had been positioned as a key part of the BBC's youth-oriented programming. "Cain, a former professional footballer being cultivated by the BBC to appeal to young male viewers, allegedly referred to women as 'slags,' 'sluts,' and 'bitches' in posts from 2015." The BBC's decision underscores the ongoing scrutiny of public figures' past social media activity, particularly when it conflicts with the corporation's editorial standards. The Guardian's reporting prompted the BBC to act, and the series will not move forward as planned.

3 sources · 19 Jun

entertainmentmedia controversymisconduct

Voice of Disney's Lilo and 'The Ring' Star Daveigh Chase Dies at 35

Daveigh Chase, the actress who voiced the beloved title character in Disney's "Lilo & Stitch" and played the haunting Samara Morgan in the horror film "The Ring," has died at age 35. According to TVLine and other outlets, her boyfriend Roy Hernandez confirmed her death to TMZ, revealing that she died from complications of meningitis and a blood infection. Chase had been hospitalized in Los Angeles after being diagnosed with meningitis and serious blood infections, Deadline reported. Hernandez had recently set up a GoFundMe page to cover her medical expenses. Variety noted that she was initially hospitalized for malnutrition before the meningitis diagnosis emerged. "Her boyfriend reported she died from meningitis and a blood infection after being hospitalized for malnutrition in Los Angeles." The actress won an Annie Award for her voice work as Lilo Pelekai in the 2002 animated film, a role she reprised in the direct-to-video sequels and the television series, according to Variety. Her performance as the eerie Samara in "The Ring" also left a lasting mark on horror cinema, with The A.V. Club noting that her career spanned from 2001 to 2010 and included work on influential horror movies, Studio Ghibli translations, and Disney animation. Decider reported that Chase's dual legacy as both a Disney icon and a horror star made her a unique figure in early 2000s pop culture. Her passing at such a young age has prompted an outpouring of tributes from fans who grew up with her performances, cementing her place in both family entertainment and genre film history.

5 sources · 19 Jun

entertainmentnews

Programming

Chinese AI Lab Z.ai Releases Open-Source GLM-5.2 With Million-Token Context Window

Chinese AI lab Z.ai has released GLM-5.2, a 753-billion parameter Mixture-of-Experts model with a 1-million-token context window, now available under an MIT open-source license. The model, which features 40 active parameters and is text-only input, was first made available to coding plan subscribers on June 13th before a public release on June 16th, according to Simon Willison's blog. "Z.ai launches GLM-5.2 with 1-million-token context window; MIT-licensed release coming next week" Hacker News reported that initial access was limited to subscribers of the GLM Coding Plan, while the broader MIT-licensed release is expected next week. This marks a significant leap from the previous GLM-5.1 model, which had a 200,000-token context window. The release is positioned as a direct response to US export restrictions targeting Anthropic models, according to Pandaily. The outlet noted that the move showcases "China's push for AI self-reliance and open-source alternatives amid escalating tech tensions between the US and China." "The key challenge addressed is making long context engineering-usable by ensuring reliability under real engineering pressure, rather than simply accepting more tokens." Hugging Face highlighted that GLM-5.2 is designed specifically for long-horizon tasks, delivering substantial improvements over GLM-5.1 by maintaining quality across long, complex coding-agent trajectories. The model's massive context window could enable processing of extremely long documents or codebases in a single pass, as noted by Hacker News. Simon Willison's blog reported strong community buzz around the release, underscoring the model's potential impact on open-source AI development. With its MIT license and million-token capability, GLM-5.2 represents a notable advancement in making long-context AI models accessible to developers worldwide.

4 sources · 19 Jun

programmingtechnology

Open-weights AI model GLM-5.2 tops Intelligence Index with score of 51

Z AI's GLM-5.2 has claimed the top spot among open-weights models on the Artificial Analysis Intelligence Index, scoring 51 and surpassing rivals like MiniMax-M3 and DeepSeek V4 Pro, according to Hacker News. The model, which shares the same architecture as GLM-5.1 with 744 billion total parameters and 40 billion active parameters, improved by 11 points over its predecessor. "GLM-5.2 becomes top open weights AI model on Intelligence Index with score of 51" This leap in performance underscores Z AI's focus on advancing scientific reasoning and other evaluation benchmarks, as reported by eicker.news. The model is available through Z AI's first-party API and third-party providers, maintaining similar pricing to GLM-5.1 at $1.4 per 1 million input tokens, $4.4 per 1 million output tokens, and $0.26 per 1 million cache hit tokens, according to Hacker News. "The model shows notable improvements in scientific reasoning and other evaluations" With a score of 51, GLM-5.2 outperforms competitors MiniMax-M3 and DeepSeek V4 Pro, both of which scored 44, as noted by both outlets. This positions Z AI as a strong contender in the open-weights AI space, offering a powerful model at competitive pricing.

2 sources · 19 Jun

ai modelsopen weightsprogramming

Microsoft Build 2025 Pushes AI Agents, Linux Tools, and a New OS for Gadgets

Microsoft's Build 2025 conference in San Francisco this week centered on winning back developers with a mix of AI-focused announcements and unexpected platform shifts. The Verge reported that the event featured Project Solara, an operating system designed for AI agent gadgets, alongside deeper Linux integration in Windows and a new mini Surface dev box. The Register noted that the conference showcased a blend of agentic AI and traditional developer tools, with a balanced view of both the aspirational nature and potential concerns around these developments. "Project Solara AI devices, Coreutils for Windows, and developer tools unveiled" The Register highlighted that Project Solara targets future devices built for AI agents rather than traditional apps, marking a significant pivot in Microsoft's hardware strategy. The Verge added that the conference also covered AI Copilot tools and support for Nvidia RTX Spark, underscoring the company's commitment to AI-driven development. A key surprise was the introduction of Unix-style Coreutils for Windows, as reported by The Register. This move, along with Execution Containers for safe AI agent execution, aims to improve the developer experience on Windows. The Verge confirmed that Windows 11 improvements and new AI models were also part of the agenda, signaling Microsoft's effort to bridge the gap between its ecosystem and the broader open-source community. The event also featured the Surface RTX Spark Dev Box, a mini developer machine, and air-gapped GitHub capabilities, according to The Register. The Verge noted that these announcements are part of a broader push to make Windows more appealing to developers who have increasingly turned to Linux and macOS. With Build 2025, Microsoft appears to be betting that a combination of AI-native hardware and cross-platform tools can reverse that trend.

2 sources · 19 Jun

aideveloper toolsprogramming
Tools

Feed Delivery, à la Bagel

Four ways to find feeds: search, browse, import, or build.

Mr Bagel
Web Search
Use the site search to find feeds from any site.
API
JSON API for feeds, entries, and search. Free key required.
MCP
Connect AI agents to FeedBagel via the MCP server.
Directory
Browse curated and categorised feeds.