Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Google

AI overview

Sign in and the AI will write an overview from our coverage.

Headlines · 932

  1. AI agent portals risk 'Uber-style' losses as usage grows

    Meta's personal AI agent Muse and OpenAI's dots have kicked off a contest for the "AI agent portal," joined by Google's Gemini Spark and xAI's Grok Bot. The analysis argues these services give each user a dedicated cloud computer and run long reasoning tasks, so more users and more delegated tasks mean higher cloud and token costs, without the ad-supported offsetting that search or social networks enjoy — an "Uber-style subsidy." OpenAI adjusted subscription pricing alongside dots, adding a $500-a-month tier and changing usage limits on the $200 Pro plan, with a sharper focus on enterprise buyers, while Meta has floated taking a cut of agent-driven transactions.

    科技新报 · 🔥 0
  2. Free Gemini app users limited to 'Auto' model picker as thinking levels roll out

    Google is rolling out a change to the Gemini app for users on the free plan: those without a plan can no longer switch models, and the only option in the model picker is "Auto" with "Gemini 3 models." Google says most prompts will go to the fastest model, Flash-Lite, while prompts needing deeper reasoning may go to Flash or Pro; if smart model selection is off, all chats for users without a plan use Flash-Lite. New thinking levels (Low, Medium, High) replace the previous "Extended" option and are available to all subscribers, matching Google Antigravity and AI Studio; per 9to5Google, Google's support documentation referred to the free-tier model as (3.5) Flash-Lite, and Google AI Plus users will soon be limited to Flash-Lite and (3.8) Flash.

    9to5Google · 🔥 0
  3. Google releases Nano Banana 2.1 with 4K output, better Chinese text, halved API prices

    Google released Nano Banana 2.1, a new image-generation model with upgrades in visual design, mask editing, subject consistency and realism, native 4K output and noticeably improved Chinese text rendering. Citing the Arena leaderboard, Chinese tech media report an average gain of about 61.7 points over Nano Banana 2 across text-to-image, multi-image editing and single-image editing, with text-to-image ranking fourth behind GPT, while API prices were halved for 1K/2K images and cut about 25% for 4K. It is available in the Gemini app and Google AI Studio (running on Gemini 3.6 Flash, with 50 free credits daily) and via the paid gemini-nano-banana-2.1 API; some users report a possible watermark-stacking bug that degrades images after repeated edits, and comparisons suggest its overall finish still trails ChatGPT Images 2.5.

    36氪 人工智能 · 🔥 0
  4. Revolut builds own foundation models as CEO eyes Meta-scale ambition

    Revolut CEO Nik Storonsky said at the Wave by Vento event in Turin that the company already trains its own foundation models on transaction data, along with its own large language models and data centres. He said AI agents will be able to shop for customers using Revolut's secure card details, that the company plans AI advice on spending, stocks and funds, and that it will open its platform so third-party agents can talk to its own agents through MCP, with every action authenticated by the customer. Revolut previously partnered with OpenAI to bundle ChatGPT Go into its subscriptions.

    The Next Web · 🔥 0
  5. 30-step image edit test: GPT Image 2.5 Sunburst drifts most, Ideogram 4.5 holds up

    Artificial Analysis ran the same photo through four frontier image-editing models for 30 consecutive edits (light the fire, add a sofa, repaint the walls, swap day for twilight, and more) to compare multi-turn consistency. It says Ideogram 4.5 and FLUX 3 edit locally, leaving 95% or more of the image essentially untouched on small changes such as adding a vase of tulips, while leaderboard-topper GPT Image 2.5 Sunburst re-renders most of the image on every edit, keeping only about a fifth unchanged so colour and detail shifts compound over turns. Nano Banana 2.1 sits in between: edits stay local but the rest of the image shifts slightly and gradually darkens. The group notes its leaderboard scores single edits, so this was its own test of what happens when 30 changes stack up.

    ArtificialAnlys (X) · 🔥 0
  6. Apple acqui-hires AI audio startup Huxe, licenses its IP

    The EU's Digital Markets Act transparency database shows Apple has struck a deal with AI audio startup Huxe giving it the right to make employment offers to certain Huxe employees and a non-exclusive license to Huxe's intellectual property. It is an acqui-hire and licensing arrangement rather than an acquisition of the company; the filing does not disclose the financial terms, which employees received offers, or whether any accepted. 9to5Mac and TechCrunch say Apple notified the European Commission on June 9, 2026, while PingWest reported the notification date as June 2025. Huxe was founded by three former Google NotebookLM team members — Raiza Martin, Jason Spielman and Stephen Hughes — launched invite-only in June 2025, opened to the public that September alongside a $4.6 million funding round, and generated daily audio briefings from email and calendar plus podcast-style conversations; it announced it was shutting down in May 2026 and ended service on May 28. Reports describe the deal as at least Apple's fourth publicly disclosed AI-related transaction this year, though Apple has said nothing about how or whether the technology will be used.

    IT之家 AI · 🔥 0
  7. This Week in AI: more capability, more responsibility

    O'Reilly's This Week in AI roundup says AI systems are gaining autonomy faster than oversight is settling: the Trump administration announced a voluntary, non-binding agreement with major AI companies covering internal safety monitoring, external audits and independent board reviews, while the FTC opened an investigation into OpenAI, Anthropic and others over consumer risks. On the technical side, OpenAI released Dots, a proactive assistant that retains context, works across apps and pursues multiple goals without prompts, and Google says Gemini 4 Argon can generate up to a million output tokens in a single response; world-model work includes AMD-acquired World Labs, Worldmodeldata's licence for nearly 1 million hours of gameplay video paired with player actions, and Physis-Lang from NVIDIA, MIT and Oxford researchers. In applied use, Anthropic's Claude is helping Ebola outbreak response teams in the DRC organize daily reports, compare forecasting models and analyze genomic information, while a Mayo Clinic model identifies patterns linked to elevated pancreatic cancer risk years before diagnosis.

    O’Reilly Radar · 🔥 0
  8. Pixel Watch 4 update brings Google's AI watch face generator

    A recent Pixel Watch 4 update (build 4.2.12.989940186) adds the AI watch face generator that debuted on the Pixel Watch 5, letting users type a prompt in the Pixel Watch app and have Gemini Intelligence generate a watch face, according to users on Reddit. The update also includes Pixel Watch 5 stock faces — Weave, Modular, Reveal and Chronograph — plus the Steph Curry dial. The rollout is said to be gradual.

    9to5Google · 🔥 0
  9. OpenAI's $70B vs $50B ARR gap is mostly a measurement issue

    A 36Kr-republished analysis by Haitian Invest argues markets overreacted to the reported drop in OpenAI's ARR from about $70B to $50B: the $70B figure is closer to a gross-revenue basis including Microsoft's cut and cloud-channel distribution, while the $50B comes from the Financial Times citing investor-facing net revenue, leaving roughly a 10% (about $6B) gap after adjustments. The piece says OpenAI's recent monthly growth has averaged about 10% versus about 3% for Anthropic, and cites an a16z report that roughly half of US users have tried AI, about a quarter use it daily and under 3% pay, while warning that Co-Working use cases are harder to validate than coding.

    36氪 人工智能 · 🔥 0
  10. EU's ENISA and Joint Research Centre test Chinese open-weight AI models

    ENISA, the EU's cybersecurity agency, and the European Commission's Joint Research Centre have begun testing open-weight AI models from China, Politico reported, with three EU officials confirming the work — the first time either body has spoken about it. Commission spokesperson Maciej Berestecki said the Chinese models sit within a wider JRC study that also covers leading US and European systems, looking at capabilities, shortfalls and risks, and are run on the centre's own secure servers without special access or licences sought from the developers. ENISA executive director Juhan Lepassaar said the agency is testing a couple of Chinese open-weight models; both bodies already have access to advanced Anthropic, OpenAI and Google systems and are jointly building an AI testing platform due by the end of this year under the Commission's July AI and cybersecurity plan.

    The Next Web · 🔥 0
  11. Spirit privacy ombudsman backs $10M sale of deidentified passenger data to Google

    Consumer privacy ombudsman Lucy Thomson recommended in a 5 October report that Spirit Airlines be allowed to sell deidentified passenger data to Google, which won the auction with a $10m bid, according to Business Insider. The new report says booking and transaction records, refunds, inflight purchases and WiFi sales are excluded from the deal and will not reach Google "in any form"; Spirit has hired Tonic.ai to deidentify the remaining unstructured data, including corporate email that may hold personal details. A court hearing on the sale is set for 14 October, and the court has the final say. Google has said it is not buying personal information and that the data would help improve its products and AI models.

    The Next Web · 🔥 0
  12. Google AI Studio Build tests unreleased Ultra mode

    TestingCatalog spotted an unreleased Ultra option in Google AI Studio Build's mode dropdown, alongside Plan, Build and the previously discovered Security mode; its description says it uses “advanced skills and tools,” but specific capabilities remain unclear. In the outlet's early tests, generations through this flow were quite comprehensive even while using Gemini 3.8 Flash, and it inferred that Ultra devotes more thinking effort to app generation than the standard flow. Whether the mode is connected to Google's announced Gemini 4 Argon or requires an Ultra subscription remains unconfirmed.

    TestingCatalog · 🔥 0
  13. Trump awards top US science medals to Musk, Huang, Su and Brin, with $2.4B AI pledge

    At the "Science: A New Golden Age" summit in Washington, President Trump presented the National Medal of Science to Elon Musk, Jensen Huang, Lisa Su and Sergey Brin, and the National Medal of Technology and Innovation to Michael Dell and Satya Nadella. Huang's citation credits him with turning the GPU into a general-purpose computing architecture, and the citations repeatedly use the administration's term "super intelligence" for AI, according to 36Kr. The White House also announced that 11 tech companies pledged a combined $2.4 billion in AI tools and compute for the Genesis Mission, including $1 billion from Nvidia, $500 million from AMD, $200 million from OpenAI and $150 million each from Anthropic and Google.

    36氪 人工智能 · 🔥 0
  14. Google moves free Gemini users to Flash-Lite, launching Nano Banana 2.1

    Google is narrowing the models available in the Gemini app for free personal accounts to Gemini Flash-Lite only, down from three. The cheapest AI Plus plan keeps Flash-Lite and Flash but loses Pro, while AI Pro and AI Ultra subscribers retain all three models; Google says affected users will get an email about when the change applies. Three days earlier Google released Nano Banana 2.1, an image model built on Gemini 3.6 Flash, which scored 1050 on overall preference in Google's text-to-image human preference tests versus 990 for Nano Banana 2. On the Gemini API, a 1K image costs $0.0336, down from $0.067, while input rises to $1.50 per million tokens from $0.50 with no free tier.

    The Next Web · 🔥 0
  15. School district bans AI in all grades weeks after Gemini pilot

    The Frederick County school board voted 6-1 to bar AI use from pre-K through 12th grade, with limited exceptions for lessons about the technology and accommodations for students with disabilities. The move reverses a decision made this school year to let middle and high school students use Google Gemini with teacher permission; an official said Gemini was the only tool evaluated that met the state's safety and privacy standards. The local teachers' union president said some educators are returning to paper-and-pencil assignments over concerns students would use AI to complete work, while some parents worry the broad ban leaves students less prepared for college.

    TechSpot · 🔥 0
  16. Google Playground hands-on: AI game generation impresses, but isn't yet fun

    A TechRadar reporter used Google's new Playground experiment to build two playable browser games from natural-language prompts alone: a fire-breathing-dragon platformer and a satirical fantasy RPG about bureaucratic adventuring. He said Playground exceeded expectations in reading his gameplay intent and generating art and music, sharply lowering the technical barrier between an idea and a playable prototype. Still, he found the results lacked the pacing, distinct characters and consequential choices of hand-crafted games.

    TechRadar · 🔥 0
  17. AI-generated bug reports overwhelm open source security, IBM says AI can help fix it

    Speaking at the Linux Foundation's Open Source Summit, IBM chief client innovation officer for enterprise security Jamie Thomas warned of a "tsunami in vulnerability disclosures," with roughly 66,000 unique entries expected in 2026 — about a fourfold increase over seven years ago — while time-to-exploit has shrunk from days to as little as 29 minutes. At the same time, AI-generated reports that are duplicated, low-quality or outright fake are swamping maintainers: curl shut down its HackerOne bug bounty, Google suspended its open source vulnerability rewards program, and Linus Torvalds called the Linux security mailing list "almost entirely unmanageable." Thomas argued the community should keep AI-powered vulnerability discovery but use AI to filter duplicates and bogus reports, assess severity and assist remediation, with efforts such as OpenSSF reducing the burden on maintainers.

    TechRadar · 🔥 0
  18. AI hardware splinters across devices, leaving the hub layer unclaimed

    A 36Kr "AI Wuzhao" analysis argues the AI hardware industry has shifted from betting on a single "next iPhone" to scattered, multi-form experimentation: glasses, pendants, earbuds, speakers, home devices and toys are all getting AI, while Humane's AI Pin has exited and Rabbit moved its OS3 to the cloud. The missing piece, the author says, is the layer above devices — identity, memory and permissions remain siloed per ecosystem, so context does not carry across hardware; cited examples include Volcengine's claim of over 1 million AIoT units running Doubao as of June 2025, JD's JoyInside across 40-plus toy brands, and Xiaomi's 1.1 billion-plus connected IoT devices with Miloco 2.0 trying to make the home the hub.

    36氪 人工智能 · 🔥 0
  19. We're putting too much faith in AI's ability to say no

    A MIT Technology Review analysis argues that teaching models to refuse harmful requests — from Anthropic's "helpful, honest, harmless" framing to the red-teaming and fine-tuning OpenAI ran ahead of ChatGPT's launch — has become the load-bearing wall of AI safety. It says researchers still describe the internal activations behind refusal only as hypotheses such as "high-dimensional polyhedral cones," so companies stack classifiers in a "Swiss cheese" model; Anthropic has said one type of classifier added 24% to its chatbots' compute costs. Because refusal is probabilistic, it can be bypassed — or turned against legitimate speech when companies or governments draw the line.

    MIT Technology Review AI · 🔥 0
  20. Hinton: humanity may have only a year or two to prepare AI oversight

    In a 55-minute interview on 《Smart Girl Dumb Questions》 released October 6, Geoffrey Hinton argued that frontier AI should have to prove its safety to independent regulators before launch, much as drugs do, rather than self-certifying, and said the window for regulation and safety preparation may be very short. He rejected the “stop AI” framing, likening oversight to a steering wheel rather than a brake, gave high marks to Anthropic CEO Dario Amodei's “Pace the Frontier” proposal, and stressed that the hard part is accepting constraints when safety rules collide with funding and product timelines. He also described using GPT to revisit the wake-sleep algorithm he worked on 30 years ago, collapsing feedback loops from about a week to minutes while keeping problem selection and judgment with the researcher.

    虎嗅 AI · 🔥 0
  21. Ai2 presents Olmo Hybrid research, outlines next Olmo

    Ai2 presented its paper "Olmo Hybrid: From Theory to Practice and Back" at COLM 2026, examining what happens when a language model combines transformer attention with linear recurrent layers. Ai2 says the model reached the same MMLU accuracy as Olmo 3 7B using 49% fewer training tokens — a result the lab reports itself, with no independent measurement supplied. Ai2 also said its next Olmo model is now in pre-training with a hybrid mixture-of-experts architecture combining attention and recurrence, and that it released Olmo-core 3, a redesigned system for training large mixture-of-experts models, ahead of the conference. The post additionally cites the open-weights AstaBrief model, the Bolmo byte-level paper's publication in Nature with extra checkpoints, and Google's recreation of the Olmo 3 7B training run in MaxText on Cloud TPUs.

    Ai2 Blog · 🔥 0
  22. Why Google is betting on RSI, starting with Gemini 4 Argon

    On Sept 30 Google released Gemini 4 Argon, the first model in its Gemini 4 series, the same day a Google DeepMind researcher posted that the model had "surpassed" him. The piece traces Google's September RSI steps — Gemini 3.8 Flash, Dream-RSI and RRSI — plus AlphaEvolve's use inside data centres, chips and AI training, arguing the company wants AI to help build its next models. Co-founder Sergey Brin is reported to be steering more resources toward RSI, as Alphabet raised its 2026 capex guidance to $195–205 billion.

    36氪 人工智能 · 🔥 0
  23. Signal's Meredith Whittaker: careless AI use is the real threat, not doom

    Signal Foundation head Meredith Whittaker told the BBC that warnings of AI turning on humanity are "not anchored in science", arguing the more immediate danger is careless use of the technology. She cited a CNN-reported incident in which a chatbot used by US intelligence fabricated information about a Chinese vessel carrying nuclear material, reportedly nearly prompting a military response, and linked Google's Project Maven to a later US missile strike on a school in Iran; Palantir said there is no evidence its software was at fault. Whittaker also said Signal would leave any market, including the UK, that forced it to weaken encryption.

    BBC Technology · 🔥 0
  24. Google app rolls out redesigned search bar with a plus icon

    Google is rolling out a redesigned home-page search bar in the Google app: it is slimmer, has a white background with a subtle bordered shadow effect, and adds a plus icon on the left for attaching images or files, creating images with Nano Banana, or switching models for AI Mode. The Nano Banana and Search Live buttons below the bar are now labeled “Create” and “Talk”. Android Authority says the new design appears with Google app version 17.65.17 and showed up on all of its test devices.

    Android Authority · 🔥 0
  25. Google AMIE study in The Lancet: AI chatbot aided clinician prep

    A study led by Google and Beth Israel Deaconess Medical Center (BIDMC), published in The Lancet's main journal, tested Google's research-grade diagnostic AI chatbot AMIE with 98 patients before their primary-care visits, with clinicians monitoring every interaction in real time. No conversation needed to be interrupted under preset safety criteria; clinicians said AI-generated summaries helped them prepare in 75% of scenarios, AI output changed their diagnostic thinking in more than half of cases, and AMIE's differential diagnoses matched physicians' final diagnoses 90% of the time. The authors say larger trials are still needed to evaluate patient-facing AI at scale.

    量子位(原生 RSS) · 🔥 0
  26. Users say Gemini as Google Assistant is 'a disaster'

    Google last month completed its migration by removing Google Assistant from all mobile devices and accounts, leaving Gemini as the default assistant. In a Reddit thread titled “Gemini as the Google Assistant is a disaster,” users report that Gemini is less reliable at everyday tasks such as making calls, setting timers and alarms, and navigating to a contact's saved address via Android Auto and Google Maps, and that it often responds with lengthy explanations instead of just acting. Some say disabling Gemini doesn't stop it from reappearing, while others can no longer find an option to switch back; these are individual user anecdotes, and the thread does not establish that every device or account behaves the same way.

    Android Authority · 🔥 0
  27. Opus 5.5 becomes the default agent brain as Google and Musk adopt it

    A commentary piece says Google added Claude Opus 5 and Sonnet 5.5 to the model picker in its Gemini Business enterprise agent at a Google Cloud event, a day after Musk said Grok Bot would outsource tasks to Opus 5.5; it frames both moves as Big Tech pivoting to Anthropic and boxing out OpenAI. The article also cites SpaceX financing documents for a prospective Anthropic compute purchase of up to about $84.5 billion, and says Anthropic's prospectus lists at least $518 billion in infrastructure commitments with six partners over the next decade. The integration details and financial figures come from second-hand accounts and social media discussion, with no confirmation from the companies named.

    Google AI on DEV · 🔥 0
  28. Google launches test satellite carrying an AI chip

    DoNews reports that a Google test satellite carrying an AI chip has been launched. The report provides no details on the chip model, orbit, launch date or mission goals, and only a headline is available, so the claim cannot be independently verified.

    DoNews · 🔥 0
  29. Innolight's triple bind as optical modules become a policy flashpoint

    An analysis piece lays out three pressures on Chinese optical-module maker Innolight: faster catch-up by peers that shortens each product generation's advantage and pushes prices down; platform players such as Nvidia reshaping system definition through NPO/CPO and locking up upstream optics (it says Nvidia invested $2bn each in Lumentum and Coherent in March 2026); and US market-access rules built on equipment authorisation and covered lists, after Reuters reported in August that the FCC was drafting restrictions and four senators proposed a bill in September targeting sensitive federal systems with a five-year transition. It cites Innolight at roughly 22.4% of the 2025 global optical-module market, with US revenue rising from 57.3% in 2025 to 61.7% in Q1 2026, and H1 2026 revenue of RMB 41.78bn with net profit of RMB 13.65bn.

    虎嗅 AI · 🔥 0
  30. OpenAI APAC public policy lead said to depart after six months

    OpenAI's most senior public policy executive in Asia-Pacific, Sanghyun Lee, is reportedly leaving the company about six months after joining, according to people familiar with the matter, as cited by Sina Finance. Based in Singapore, Lee joined OpenAI in April this year after six years at Google; the reason for his departure is not known. The move comes as the ChatGPT developer faces scrutiny from regulators in the region.

    36氪快讯 · 🔥 0

← PreviousNext →

Experience and discussion from the community

Share my Google experienceAsk about Google

design.md: A format specification for describing a visual identity to coding agents.

What problem it solves Coding agents need both exact design values and the rationale for applying them consistently. Without a structured, persistent representation of a design system, agents may miss design context or interpret visual guidance inconsistently. Key features Structured tokens and design rationale — The format combines machine-readable design tokens with human-readable Markdown ratio…

github.comhttps://github.com/google-labs-code/design.md

pi: A minimal, extensible agent harness that you can make your own.

What problem it solves Pi is designed for developers who need to adapt an agent to their own workflows rather than work within a fixed feature set. It provides interactive, automated, RPC, and SDK interfaces while leaving optional capabilities such as sub-agents and plan mode to extensions or packages. Key features Extensible agent harness — Customize the harness with extensions, skills, prompt te…

github.comhttps://github.com/earendil-works/pi

Antigravity-Manager: Multi-platform automation and multi-account matrix dispatch console for AI services.

What problem it solves Managing multiple AI service sessions and translating them into compatible API interfaces can require repetitive account switching, custom routing, and protocol-specific integration. The project addresses these tasks with a desktop application that combines account administration, protocol conversion, request scheduling, and model routing. Key features Account dashboard and …

github.comhttps://github.com/lbjlaq/Antigravity-Manager

app-store-compliance: The enterprise reference and automated guard that turns App Store and Google Play rejection into a designed out failure mode.

What problem it solves App Store and Google Play rejections can leave teams searching for the cause of a brief review message, fixing it, and resubmitting. The project addresses this with documented rejection patterns, concrete fixes, pre-submission checks, and an automated guard that can run before an upload. Key features Pre-submission project scanning — The guard scans iOS and Android projects …

github.comhttps://github.com/mjmirza/app-store-compliance

paper-orchestra: A multi-agent framework for automated AI research paper writing.

What problem it solves Researchers often need to turn unconstrained ideas and experimental logs into complete research manuscripts with literature synthesis, structured sections, refinements, and figures. PaperOrchestra addresses this gap by coordinating specialized agents across the paper-writing workflow and producing LaTeX manuscripts that follow target conference templates. Key features Multi-…

github.comhttps://github.com/google-research/paper-orchestra

assistant-ui: An open-source TypeScript/React library for building production-grade AI chat experiences.

What problem it solves Building AI chat experiences in React requires assembling chat primitives, runtime integrations, generative UI, and production UX details. The project provides a composable library that supports multiple backends, custom styling, and strongly typed APIs. Key features Composable Chat Primitives — Build a chat interface from primitives including Thread, Message, Composer, Thre…

github.comhttps://github.com/assistant-ui/assistant-ui

ax: A declarative orchestrator for running autonomous agent workloads in a cluster.

What problem it solves Autonomous agents accumulate state, require strict isolation, call model APIs and tool servers, and can continue consuming resources without supervision. AX addresses these requirements through declarative primitives for sandboxed execution, prepared workspaces, model configuration, lifecycle control, and direct inspection. Key features Declarative task manifests — Tasks, wo…

github.comhttps://github.com/google/ax

supoclip: An open-source, AI-powered video clipping tool.

What problem it solves Long-form videos need to be turned into short, platform-ready clips, but metered usage, platform branding, and server-side control can constrain the workflow. SupoClip addresses these concerns with a self-hostable, open-source pipeline that offers no-watermark output and a hosted option. Key features AI clip selection and scoring — A supported language model selects clip-wor…

github.comhttps://github.com/FujiwaraChoki/supoclip

elephant-goldfish-model: An open-source plugin for AI assistants that supports software design and review.

What problem it solves Software design often depends on context accumulated in an earlier conversation, while reviewers can miss assumptions and blind spots when they share that same context. EGM separates context-rich design work from fresh-context review so findings can be surfaced before implementation. Key features Context-aware design workflow — Explore requirements and trade-offs, save the a…

github.comhttps://github.com/google/elephant-goldfish-model

opencodex: A lightweight local gateway for using chosen models with AI coding tools.

What problem it solves AI coding tools often expect particular provider protocols, making it difficult to use the models and providers a user prefers. A local gateway can translate between supported client APIs and provider APIs while preserving capabilities such as streaming, tool calls, reasoning, and images. Key features Provider protocol translation — The gateway translates Responses, Chat Com…

github.comhttps://github.com/lidge-jun/opencodex

crowdsec: An open-source and participative security solution offering crowdsourced server detection and protection against malicious IPs.

What problem it solves Server operators need to detect malicious behavior across logs and HTTP requests, then apply remediation at the appropriate layers of their infrastructure. They also need a way to benefit from threat intelligence shared by the community. Key features Integrated Security Engine — The Security Engine combines IDS/IPS and WAF capabilities in one solution. It analyzes log source…

github.comhttps://github.com/crowdsecurity/crowdsec

avenoxbeyin: A local memory system for AI agents that keeps notes, decisions, tasks, and sources in Markdown.

What problem it solves AI clients change between sessions, making it difficult to preserve context, recover earlier decisions, and continue unfinished work. The project provides a local Markdown-based memory that records selected information and retrieves it with its source in later sessions without automatically converting the entire conversation history into knowledge. Key features Persistent Ma…

github.comhttps://github.com/avenoxai/avenoxbeyin

system_prompts_leaks: A regularly updated archive of leaked system prompts from ChatGPT, Claude, Gemini, and other AI products.

What problem it solves The hidden system instructions that shape AI product behavior are difficult to inspect and change across models, products, and versions. This repository makes collected prompts available in a structured, product-oriented archive for study and comparison. Key features Broad product coverage — The archive covers prompts from major AI providers and products, including Anthropic…

github.comhttps://github.com/asgeirtj/system_prompts_leaks

n8n: A fair-code platform to build and deploy AI agents and workflows.

What problem it solves Building AI agents and workflows often requires combining visual orchestration with custom code while connecting them to existing systems. The platform addresses these needs with model flexibility, self-hosted or cloud deployment, workflow logic, human approvals, and observability. Key features Visual and code-based workflow building — Build workflows on a visual canvas and …

github.comhttps://github.com/n8n-io/n8n

google-ads-mcp: An MCP server that interacts with the Google Ads API and provides tools and resources for LLMs and AI agents.

What problem it solves LLMs and AI agents need structured access to Google Ads account data, API metadata, and management operations. The server provides this access through MCP while adding a two-step proposal and confirmation workflow for write operations. Key features Query Google Ads accounts — Use GAQL to retrieve information about a Google Ads account, list directly accessible customers, and…

github.comhttps://github.com/googleads/google-ads-mcp

browser-use: An open-source Python library for AI browser agents that operate real browsers through a persistent Python REPL and raw Chrome DevTools Protocol.

What problem it solves AI agents need a reliable way to perform browser tasks such as finding availability, selecting options, handling CAPTCHAs, and completing forms. Developers also need flexibility to run the agent locally or through hosted infrastructure while choosing their model, browser mode, tools, and output format. Key features Persistent Python agent runtime — The model writes Python in…

github.comhttps://github.com/browser-use/browser-use

modern-web-guidance-src: Expert-curated web platform guidance that embeds modern practices and browser compatibility data into coding agents.

What problem it solves Coding agents often default to legacy patterns because their training data contains substantial amounts of legacy code, which can produce bloated JavaScript where native web platform solutions are available. They may also know that a modern API exists without having enough real-world implementation patterns to apply it correctly in production-ready code. Key features Context…

github.comhttps://github.com/GoogleChrome/modern-web-guidance-src

google-cloud-go: Go packages for Google Cloud Platform services.

What problem it solves Go applications need client libraries for interacting with Google Cloud Platform APIs. The project provides packages for those services and supports common authentication flows for local development and Google Cloud Platform environments. Key features Google Cloud API client packages — The repository contains Go packages for Google Cloud Platform services, including storage,…

github.comhttps://github.com/googleapis/google-cloud-go

stitch-skills: A collection of agent skills and plugins for Google Stitch, following the Agent Skills open standard.

What problem it solves Coding agents need a structured way to create, manage, and build from Google Stitch designs while integrating with the Stitch MCP server. The project organizes design, code generation, and supporting workflows into skills that can be installed with supported coding agents. Key features Design workflows — The design plugin covers converting frontend code into Stitch designs, …

github.comhttps://github.com/google-labs-code/stitch-skills

OpenStock: OpenStock is a free, open-source alternative to expensive market platforms.

What problem it solves Market platforms can place broad access to prices, charts, company information, watchlists, and alerts behind paid subscriptions. OpenStock addresses this by providing a community-built platform that can be hosted publicly or self-hosted with every feature. Key features Multi-market dashboard — Explore eight markets through a dashboard with an index pulse, sector heatmap, ma…

github.comhttps://github.com/Open-Dev-Society/OpenStock