Skip to content
Latest updates
QuotaTick

AI coding tools news

November 12, 2026 · 30 stories

The day's news for developers who use AI coding tools, picked and summarised from official changelogs, the media and Hacker News. Resets and limit changes are on the reset tracker.

  1. GitHub Copilot CLI v1.0.110 adds password prompts and session fixes

    GitHub Copilot CLI v1.0.110 adds experimental masked dialog support to answer git, ssh, and sudo password prompts from agent commands. It also fixes Microsoft Entra sign-in error popups, improves screen reader session announcements, and wraps /usage model rows at terminal width.

    Source · Official

  2. Kiro records service MCP tools calls in AWS CloudTrail

    Kiro now records calls to Kiro service MCP tools in AWS CloudTrail for Kiro profiles in US East (N. Virginia) (us-east-1) and Europe (Frankfurt) (eu-central-1). Security and audit teams can view who made a request, which tool was called, when it ran, and its completion status.

    Source · Official

  3. Google launches Gemini 4 Hydrogen for Pro and Ultra users

    Google launched Gemini 4 Hydrogen for faster coding and agentic workflows, featuring four thinking levels and flat API pricing. The model is rolling out broadly to Pro and Ultra users.

    Source · Media report

  4. Claude Code v2.1.323 adds Code Mode and plugin validation

    Claude Code v2.1.323 introduces an experimental Code Mode for JavaScript tool calls and adds type checking for plugin TypeScript hooks. The update includes support for CLAUDE_CODE_AUTO_MODE_MODEL, .docx file uploads in Artifacts, and a 30-second progress heartbeat for long-running hooks. It also adds an options menu for usage credit management and new warnings for Bash allow rules.

    Source · Official

  5. Muse AI adds outbound calling feature for US businesses

    Muse AI now supports placing outbound calls to businesses in the United States. This feature is available as a paid service.

    Source · Media report

  6. Google launches Gemini 4 Hydrogen in Search and Gemini app

    Google has launched the new Gemini 4 Hydrogen model, designed to handle complex, multi-part questions, typos, and specific requests. The model is available starting today in search, the Gemini app, and across all Google tiers.

    Source · Media report

  7. Google launches Gemini 4 Hydrogen coding agent with mid-task model switching

    Google launched Gemini 4 Hydrogen on Thursday alongside a new system that switches models mid-task during coding. The new system allows coding agents to finish tasks 19% faster.

    Source · Media report

  8. Nebius plans $40 billion AI infrastructure expansion

    Nebius plans to fund a $40 billion AI infrastructure expansion through customer prepayments and loans.

    Source · Media report

  9. Claude Code adds Claude Meetings beta for Team and Enterprise plans

    Claude Meetings is rolling out in beta on Team and Enterprise plans in Claude and Claude Desktop. You can add Claude to a Google Meet call as a visible participant that takes notes, answers questions, and builds artifacts such as docs, whiteboards, and slides during the call.

    Source · Official

  10. Devin Desktop update adds security fixes and several usability improvements

    Devin Desktop now runs on Electron 43.8.8, which includes Chromium security fixes and the Node.js 24.22.1 security update. The model picker no longer ignores clicks while the app refreshes account status. Images sent to GPT and Gemini models are limited to 2048 px (GPT) or 3072 px (Gemini) on the longest side. Overflow files older than 7 days are deleted at startup. Other fixes include preventing Cloud session creation in some enterprise accounts, avoiding corruption of non‑UTF‑8 files, and fixing crashes with Markdown tables in background tabs.

    Source · Official

  11. Nectar launches AI agent for creator discovery and outreach

    Nectar's new AI agent automates influencer marketing. It helps brands discover creators, manage outreach and partnerships, and track campaign results.

    Source · Media report

  12. Google Search upgrade adds Gemini 4 Hydrogen to AI Mode

    Google’s AI Mode is now powered by Gemini 4 Hydrogen, improving understanding of complex queries, surfacing relevant links and delivering richer visual responses.

    Source · Media report

  13. Antigravity adds Gemini 4 Hydrogen and Carbon to model selector

    Antigravity v2.31.2 adds Gemini 4 Hydrogen, which provides fast coding and agentic execution across four reasoning effort levels and includes Orchestrator for dynamic multi‑model planning with Gemini 4 Carbon. The release also adds Gemini 4 Carbon to the model selector when signing in with Google Cloud application default credentials, selectable at low, medium, or high thinking levels. v2.31.1 fixes Gemini Enterprise sign‑in on Windows for Workforce Identity Federation accounts that could lose the selected license.

    Source · Official

  14. Jensen Huang resigns, raising questions on Nvidia’s open‑source AI support

    Jensen Huang resigns from Nvidia. The article says his departure raises questions about Nvidia’s commitment to open‑source AI, including its Hugging Face acquisition, open‑weight models and advocacy against restrictions, as the company undergoes a leadership transition.

    Source · Media report

  15. Antigravity adds Gemini 4 Hydrogen and Orchestrator

    Antigravity announced the addition of Gemini 4 Hydrogen and Orchestrator. The announcement was posted on 12 Nov 2026.

    Source · Official

  16. United Airlines blocks AI agents from booking flights

    United Airlines said it will not allow customers to use AI agents to book flights and manage reservations.

    Source · Media report

  17. Kiro adds credit pooling support for Okta and Microsoft Entra ID

    Enterprise administrators whose Kiro profile connects directly to Okta or Microsoft Entra ID can now share Kiro credits across their team with credit pooling. New profiles that use IAM Identity Center, Okta, or Microsoft Entra ID start with credit pooling. The Kiro console’s Access management > Credit sharing section is used to add users and groups, buy seats, set usage limits, and move existing subscribers via the Switch to credit sharing wizard.

    Source · Official

  18. Claude Managed Agents now self‑hosted

    We’ve launched self‑hosted Claude Managed Agents. You run the whole Managed Agents service, not only self‑hosted sandboxes, on your own infrastructure, and you store agents, sessions, and session history there. The server sends model requests to the provider you choose, Anthropic by default. To request access, contact your Anthropic account team.

    Source · Official

  19. Google Gemini Decisions API launched

    Google introduced the Gemini Decisions API on Wednesday. The API is designed to handle routine decisions that AI applications make thousands of times. No pricing or usage limits were disclosed.

    Source · Media report

  20. Phota Labs launches AI‑powered automatic photo editing service

    Phota Labs announced a service that uses AI to automatically edit everyday photos into professional‑looking shots. The tool is intended to make photo editing easier for consumers.

    Source · Media report

  21. Jensen Huang agrees to testify before Congress

    Jensen Huang, ousted as Nvidia CEO, agreed to testify on Wednesday after senior lawmakers threatened a subpoena. The testimony follows reports of an alleged affair with a woman suspected of spying for China.

    Source · Media report

  22. Nvidia CFO Colette Kress appointed interim CEO replacing Jensen Huang

    Nvidia CFO Colette Kress is taking over as interim CEO, replacing longtime CEO Jensen Huang. The change is effective immediately. No further details on tenure or compensation are provided.

    Source · Media report

  23. Fireworks AI data breach exposes Notion customers' chat transcripts

    Fireworks AI suffered a data breach, the second in as many months. Hackers accessed chat transcripts of Notion customers.

    Source · Media report

  24. MiniMax releases open‑source multimodal “Source God” raw‑image model, native 4K, ¥0.2 per image

    MiniMax released an open‑source multimodal raw‑image model named “Source God”. The model supports native 4K resolution. Pricing is ¥0.2 per generated image.

    Source · Media report

  25. Kimi pricing page updated

    The Kimi pricing page removed the Kimi K2.7 Code model description and the Kimi K2.6 model description.

    Source · Official

  26. OpenAI commits $2 M to AI and children safety research

    OpenAI commits $2 million to research on AI and children’s safety. The work supports shared tools to measure risks and opportunities for young people.

    Source · Official

  27. Anthropic launches IPO roadshow, valuation down $600 B

    Anthropic launched its IPO roadshow. The company’s valuation has decreased by $600 billion.

    Source · Media report

  28. OpenRouter adds support for 8 languages

    OpenRouter now supports eight languages. The announcement details the translation pipeline and the associated costs. It also explains how reader reports are used for fixes.

    Source · Official

  29. OpenRouter pricing/limits page adds API key usage examples

    The OpenRouter documentation page now includes example code for retrieving the current API key using the JavaScript SDK, a fetch request, and Python requests. The snippets show how to log the key information. No pricing or quota numbers were changed.

    Source · Official

  30. GitHub Copilot CLI v1.0.110-2 adds credit‑limit enforcement and fixes

    The release adds /compact respecting the session AI credit limit to stop extra charges after the limit is reached. It also fixes label handling for incomplete responses, directory‑not‑found messages, plugin listing explanations, and Vim cursor behavior. The changes affect all users of the CLI on any platform.

    Source · Official