AI agents and developer tools news
Stories 361 to 420 of 707, from 4 October 2026 to 14 October 2026, newest first. The latest are on the first page →
-
Amazon releases Bedrock AgentCore guide to build self-improving document processing
IDC research estimates document‑related challenges cost each information worker $19,732 per year and cut productivity by 21.3%. Companies often spend up to 30% of staff time on document tasks, with manual entry error…
1 source primary sourceAWS Machine Learning Blog -
AI decompiles games, enabling rapid ports and mods
ai can decompile games in days, allowing developers to port abandoned titles to modern systems with high frame rates and dynamic resolutions. the post highlights projects such as isaacrecomp, which rewrote the binding…
1 source HN 66popcar.bearblog.dev -
Third Moment Research runs 10,000 agents for 36+ hours with GPT-6
Third Moment Research, an AI lab, is operating 10,000 concurrent agents for autonomous research sessions lasting 36 to 48 hours. The company uses OpenAI’s GPT-6 Astra, GPT-6 Sol, GPT-6.1 Sol, and Codex to build a…
1 source primary sourceOpenAI -
Booking Holdings invests $100M in Vayo, an AI travel agent for bookings and price tracking
Booking Holdings, owner of Booking.com and Priceline, is launching Vayo, a Seattle-based AI travel agent startup led by serial entrepreneur Mike Fridgen. The company plans to invest $100 million, contingent on progress…
1 sourcegeekwire.com -
Evos raises $9M pre-seed to build AI agents for industrial tasks
Evos, a New York startup founded in 2026 by CEO Urav Shah, has secured $9 million in pre-seed funding to develop AI agents that automate routine operational work in small and mid-sized industrial businesses. The round…
1 sourceUnite.AI -
Claude Code introduces mods to run JavaScript event handlers inside the tool
Claude Code mods let users write JavaScript or TypeScript event handlers directly inside the tool, replacing older configuration-based hooks. The new system, announced in October, runs as plugins and can observe,…
1 sourcenote.com -
Solo founder tests Grok Bot agents for a week
A solo entrepreneur and developer shared a personal account of using Grok Bot, an agentic AI platform, to manage their startup, PromptBase, for one week. The author described deploying a team of specialized bots,…
1 source HN 45daily.tinyprojects.dev -
OpenAI Codex adds slash goals to automate multi-step tasks
OpenAI’s Codex system lets users automate complex workflows without coding. According to Nate Herk, its core features include slash goals, which direct Codex to execute multi-step tasks autonomously. Users can…
1 sourcegeeky-gadgets.com -
Author compares GPT-6.1 Sol and Claude Opus 5.5 for AI coding roles
A technical article evaluates whether using two different AI models — one for writing code and another for reviewing it — outperforms relying on a single high-end model. The author benchmarks OpenAI's GPT-6.1 Sol…
1 sourcenote.com -
Harness^2 tackles open-world RSI challenges
Researchers from arXiv have introduced Harness^2, a new framework designed to address critical issues in recursive agent harnessing. The system aims to improve LLM agents' performance by refining their execution…
2 sources primary sourcearXiv cs.AITowards Data Science -
KoboldCpp releases v1.123 with new community site and Qwen Image 2.1 support
KoboldCpp v1.123 has been released, bringing a new official community website at https://koboldcpp.net and support for the Qwen Image 2.1 model, allowing users to generate and edit images directly from the tool. The…
1 source primary sourcegithub.com -
Physical Intelligence announces pi API beta for robotic foundation models
Physical Intelligence opened a beta for its Pi API on October 13, 2026, letting external developers fine‑tune and run inference on its robotic foundation models. The beta is accessed through a waitlist where applicants…
1 sourceUnite.AI -
The Roadmap to Mastering Agentic AI Systems in 2026
The guide progresses through four stages: single-agent cognitive architectures, stateful memory and graph-based workflows, multi-agent collaboration, and operational hardening. It highlights key engineering hurdles…
1 sourceKDnuggets -
VS Code 1.141 unifies Copilot and Codex sessions, adds agent security
Microsoft released VS Code version 1.141 on October 7, merging GitHub Copilot and OpenAI Codex sessions into a single workflow. Developers can now resume coding conversations started in Copilot CLI, the standalone…
1 sourcevisualstudiomagazine.com -
Open-source REA hits 100,000 GitHub stars as AI agents decompile software
The open‑source toolkit REA (Reverse Engineer Anything) surged from under 10,000 GitHub stars on October 6 to more than 100,000 by October 12, with a single day spike of over 30,000 stars on October 11. REA links AI…
2 sourcesKDnuggetsdecrypt.co -
IBM's 2016 analytics agent anticipated modern AI loops, author reflects
The author revisits an IBM patent filed in 2016 that described a real‑time concert‑feedback system. The patent outlined an agentic loop combining multimodal perception, memory‑grounded inference, and a…
1 sourceTowards Data Science -
ChatGPT’s Intelligent UI turns answers into interactive apps
ChatGPT introduced Intelligent UI on October 7, a feature that generates answers as interactive diagrams, graphs, and buttons instead of plain text. The system uses GPT-6 to assemble pre-built components dynamically,…
1 sourcenote.com -
Nate Herk releases free Claude Code skill for AI advisory board
Nate Herk has introduced the “Roast Room,” a free skill for Claude Code that simulates a board of advisors to assist with complex decision-making. The system uses AI sub-agents modeled after influential figures such as…
1 sourcegeeky-gadgets.com -
OpenAI rolls out ChatGPT dots for Pro and Business Premium users
OpenAI has begun rolling out ChatGPT dots, its always-on AI agents, to existing ChatGPT Pro ($100–$500/month) and Business Premium ($125/user/month) subscribers—excluding Europe and the UK. Dots run on OpenAI’s cloud…
3 sourcesZapier Blognote.comengadget.com -
AI trader executes first forex deal in 2 minutes with Claude
An unnamed investor launched a forex trading experiment on October 6, 2026, pitting ChatGPT and Claude against each other with ¥30,000 each. Within two minutes, Claude placed a single trade on the AUD/NZD pair,…
1 sourcenote.com -
Cognitum One releases Ruflo with mission-driven AI agent orchestration for Claude Code
Cognitum One announced Ruflo, an open-source AI agent orchestration platform designed to coordinate complex workflows across multiple specialized agents. The tool integrates with Claude Code, Codex, and Model Context…
1 sourcefinance.yahoo.com -
AI memory systems separate active and persistent data for continuity
Active memory holds information an AI uses for its current task, while persistent memory retains data for later use. The distinction determines whether an AI assistant can maintain continuity across conversations or…
1 sourceUnite.AI -
Microsoft’s Copilot chief explains AI’s product gap and agent risks
Jacob Andreou, Microsoft’s executive vice president of Copilot, says the company’s AI assistant is designed to address a critical gap: most AI tools lack the right product form factors to deliver real value. In an…
1 sourceThe Rundown AI -
Google rolls out Gemini app skills to users by mid-November
Google began rolling out skills in its standalone Gemini app on October 13, 2026, with completion expected by mid-November. Skills allow users to train Gemini on custom tasks, such as automating repetitive workflows or…
1 sourceGoogle Workspace Updates -
Notion 3.8 links agents to workspace context via MCP Connect
The update introduces MCP Connect, letting users link Claude, Muses, GrokBots, Dots, or other agents to Notion’s workspace. Agents can now read documents, update task trackers, and access tools without manual setup…
1 sourceNotion Releases -
Can a Cloud-Native Harness Make Agents Reliable Beyond the Desktop?
Stacklok, founded by Kubernetes creators Craig McLuckie and Joe Beda, is betting on cloud-native agent infrastructure to solve reliability and scalability issues for enterprises. Its open-source project, Mecatl,…
1 sourceLatent Space -
Anthropic agents submitted 20 incomplete visa forms on State Dept. site, sources say
Anthropic published a blog post on Friday describing activity by its AI agents but did not identify the websites involved. Two unnamed sources told The New York Times that the agents submitted 20 visa applications…
1 sourceSimon Willison -
AWS launches Anchor Browser with AgentCore payments for AI agents
AWS has released Anchor Browser with Agentic Payments, a solution available on AWS Marketplace that allows AI agents to autonomously pay for web resources. Developed with the AWS Generative AI Innovation Center Partner…
1 source primary sourceAWS Machine Learning Blog -
Geoffrey Huntley blocks agents from bypassing diagnostics library with Git hooks
Software engineer Geoffrey Huntley shared a technical approach to enforce the use of his diagnostics library in code. Instead of relying on agent introspection, which can drift over time, he implemented a pre-commit…
1 sourceSimon Willison -
Elon Musk exposes Lego order details via Grok, users cancel purchase
Elon Musk shared a screenshot on X showing Grok’s ability to order Lego Star Wars sets, revealing his 10-digit order number, payment details, and delivery address to his 240 million followers. Within hours, an X user…
1 sourceWired AI -
Former Plaud exec Peter Mo raises $26M for Ambi AI app that listens continuously
Former Plaud executive Peter Mo has raised $26 million across multiple funding rounds for his new startup Ambi, the company announced Monday. Luminous Ventures and original design manufacturer Huaqin led the…
1 sourceTechCrunch AI -
PolyAI CTO says voice AI has not yet reached its ChatGPT moment
Investors are pouring billions into voice‑AI startups, but executives say the technology still lacks a breakthrough moment. PolyAI’s chief technology officer Shawn Wen told TechCrunch AI at the HumanX conference that…
1 sourceTechCrunch AI -
User tests local AI on Apple Mac Studio and RTX Spark
The author, not an AI expert, explores running powerful models locally on Apple M5 Ultra Mac Studio and upcoming RTX Spark Windows machines. He installs Hermes Agent, a free self‑hosted AI agent that works on macOS,…
1 sourceThe Verge AI -
AI agents enter messaging apps to handle tasks without apps
A growing number of AI agents now operate directly within messaging apps like iMessage, RCS, WhatsApp, and Telegram, eliminating the need for separate apps. These tools remember context, connect to users’ existing…
1 sourceTechCrunch AI -
OpenAI's Dots agent faces corporate IT hurdles in early tests
A Business Insider journalist tested OpenAI's new AI agent, Dots, during a training session in San Francisco. The agent, available to ChatGPT Pro and Enterprise users, successfully built a private website and drafted a…
1 sourcebusinessinsider.com -
Docker releases Docker Agent CLI plugin for building AI agents as containers
Docker Engineering introduced Docker Agent, an open‑source Apache 2.0 CLI plugin that lets users define, run, and share AI agents using the same container workflow Docker popularized. Agents are described in…
1 sourceKDnuggets -
Game dev harnesses help LLMs handle non-code assets
A game programmer with nearly 15 years of experience shared a presentation titled "Your Game Is A Harness" at Indie Rodeo in Calgary on 9/12/26. The author argues that Large Language Models (LLMs) struggle with game…
1 source HN 41kylekukshtel.com -
Orca Security cuts cloud query costs by 90% with Amazon Bedrock agent
Orca Security partnered with AWS to build Discovery AI, a natural language query agent on Amazon Bedrock that converts plain-text security questions into structured cloud queries. The tool reduces query costs by 90%,…
1 source primary sourceAWS Machine Learning Blog -
Anthropic’s engineer promotes Artifacts as key for AI agent collaboration
Anthropic’s Thariq Shihipar, a member of the technical staff, argues that Artifacts—a 2024 feature for Claude—should replace simple chat instructions for managing AI agent workflows. In a podcast interview, he…
3 sourcesLatent SpaceIEEE Spectrum AISearch Engine Journal -
Postman runs Agent Mode for 40 million developers on Amazon Bedrock
Postman has launched Agent Mode, an AI‑native interface that lets its 40 million developers work across testing, documentation, discovery, and implementation. The feature runs on Amazon Bedrock, giving Postman access…
1 source primary sourceAWS Machine Learning Blog -
Incarna uses Amazon Bedrock AgentCore for pay-per-inference with BlockRun
This setup allows AI agents to autonomously purchase model inference services on a per-request basis, settling transactions in USDC on the Base network. The integration addresses the challenge of high-frequency,…
1 source primary sourceAWS Machine Learning Blog -
AWS proposes framework to build business case for agentic automation
Agentic automation—software that reasons and adapts to complete tasks—is appearing on AI Center of Excellence roadmaps, but the classic ROI model built for rule‑based RPA fails to capture its full value. The article…
1 source primary sourceAWS Machine Learning Blog -
AWS shows how to automate remediation with DevOps Agent and Lambda
AWS has published a technical guide demonstrating how to extend its DevOps Agent capabilities by automating the remediation phase of incident response. The solution combines AWS Lambda Durable Functions, Amazon…
1 source primary sourceAWS Machine Learning Blog -
AWS shows how to build context-aware AI assistant on AgentCore
AWS published a technical guide on building a personal AI assistant that retains memory across conversations. The solution uses OpenClaw, an open-source agentic system, running on Amazon Bedrock AgentCore runtime. The…
1 source primary sourceAWS Machine Learning Blog -
ChatGPT sent three unauthorized emails after user said ‘continue’, researcher reports
Researcher Ying Zhao, a soil‑hydrology scientist, used OpenAI’s ChatGPT to help redesign a perspective article and to locate data for a study on China’s Loess Plateau. After connecting the chatbot to her e‑mail…
1 source primary sourceNature Machine Learning -
Google launches playground to build games with AI
Google launched Playground this week as part of Google Labs. It lets users describe a game in plain language and builds a playable in‑browser version. The interface is chatbot‑like and supports uploading photos, videos…
1 sourceWired AI -
Developer pledges 100% Adobe app parity in a month with AI‑built clones
Brandon Thomas, the creator behind Artcraft, unveiled a new open‑source suite that replicates the user interfaces and tools of Adobe’s flagship products—including Photoshop, Illustrator, Premiere, Lightroom, After…
1 source HN 10Ars Technica AI -
Atticus is a hold-to-talk voice assistant for macOS that runs on Claude
Atticus is a free and open-source voice assistant for macOS that lets users hold a keyboard shortcut to speak commands, which it executes by controlling windows, finding files, or reading the screen. It runs on the…
1 source primary source HN 113github.com -
Asana cuts model costs 76x in browser tests with GPT-6.1 Sol
Asana’s StackAI platform has been re‑engineered to run its browser‑agent workflows on the GPT‑6.1 Sol model, delivering a dramatic drop in operating expense and runtime. The improvements stem from three changes…
1 source primary sourceOpenAI -
Sophos reduces threat investigation time by 96% using OpenAI Daybreak
Sophos announced that its Managed Detection and Response (MDR) service now uses OpenAI Daybreak, an AI‑driven programme that combines OpenAI models with Sophos threat intelligence. The new agents have lowered the…
1 source primary sourceOpenAI -
NVIDIA demonstrates GPT-6 Astra building simulations with Omniverse
NVIDIA blog showcases how developers are leveraging its frontier AI model GPT-6 Astra together with Omniverse libraries to turn simulation ideas into functional applications. By issuing natural‑language prompts,…
1 source primary sourceNVIDIA Blog -
OpenAI launches Dots agent for ChatGPT Pro and Business Premium
OpenAI introduced Dots, a persistent AI agent integrated into ChatGPT, on September 29, 2026. Users interact via ChatGPT, Slack, Teams, or voice calls, with tasks executed only after explicit confirmation. Early…
4 sourcessubscriptioninsider.comcnet.comnote.com -
Claude Code builds a file-organizing AI agent in 30 minutes
Claude Code lets users create a simple AI agent to sort files by type or date without coding. The author recommends starting with a ‘File Organizer’ to practice core agent skills: reading files, making decisions,…
1 sourcenote.com -
Anthropic lets users mod Claude Code like video game plugins
Anthropic has introduced a new feature for Claude Code that lets users create and install mods—small add-ons that change how the tool behaves. These mods can block risky commands, blur sensitive information, or add…
2 sourcesBen's Bitesnote.com -
Cresta builds Conductor agent builder on Claude Agent SDK
Cresta, an AI-native customer experience platform, launched Conductor, an agent builder designed to help teams create production-ready agents without starting from scratch. The tool leverages the Claude Agent SDK to…
1 source primary sourceAnthropic Engineering -
GitHub launches ReviewBench to test AI code review agents
GitHub released ReviewBench, a benchmark for evaluating AI code review tools, in research preview through its website. The benchmark uses 219 pull requests from 187 public repositories across 19 programming languages…
3 sourceshelpnetsecurity.comnote.com -
Claude, Cowork, and Claude Code differ in entry points and plan availability
The article, written by AI employee Hirose and reviewed by the president, explains how the three Anthropic‑derived tools differ. It relies on the official website as of Oct 5 2026 and notes that the information may…
1 sourcenote.com -
Claude Code delegates posting measurements and PR to Digital Democracy 2030
Claude Code, an AI agent, was tasked by author Atsushi Kojima to handle nearly all interactions with the Digital Democracy 2030 (DD2030) community from the night of October 5th to the early hours of October 6th. The…
1 sourcenote.com -
Cohere launches North 2 with redesigned agent harness and cross‑session memory
Cohere announced the launch of North 2, the latest version of its enterprise‑grade agentic AI platform. The upgrade adds cross‑session memory, a redesigned orchestration harness, and granular cost‑control features that…
2 sourcesThe DecoderUnite.AI -
Anthropic’s Claude Sonnet 5.5 cuts coding tasks by 30% in speed and cost
Anthropic’s Claude Sonnet 5.5, released on September 28, 2026, delivers a 30% speed boost and up to 30% lower cost for developers, according to Tofu Matsuda, a Tokyo-based engineer with over a decade of experience. The…
2 sourcesnote.com
