AI TrackerCoding news,
for builders.
The latest coding news from every AI provider: launches, price changes and retirements, summarized with what they mean for you.
Anthropic finds no regression behind Claude Code usage complaints, and fixes a few token-burn bugs
Boris Cherny of the Claude Code team posted on X on October 28 that the team went through all the session IDs users shared after complaints that Claude Code was using up limits faster. He says they found a few smaller issues that made some sessions burn through tokens faster than they should, and that fixes are rolling out now.Command Code revokes all API keys after a security incident tied to a distillation attack
Command Code, a coding agent, posted on X on October 28 that earlier that day it "detected and stopped a security incident linked to a large-scale distillation attack targeting Anthropic and OpenAI models." This is the company's own account.Particle-1, a model post-trained on DeepSeek V4.1 Flash, is now on Vercel AI Gateway
Vercel says Particle-1 (particle/particle-1) is now available on AI Gateway, its unified endpoint for model providers. Vercel says the model is post-trained on DeepSeek V4.1 Flash and completes engineering tasks with fewer tokens, and suggests it for debugging, planning code changes and breaking down multi-step problems.Vercel launches Vercel CI in beta on all plans
Vercel introduced Vercel CI, a continuous integration product now in beta on all plans. It skips tasks intelligently using Turborepo, and is optimized for agents through the Vercel MCP server and CLI.Vercel's open-source agent harness fx is now a built-in provider in T3 Code
Vercel says fx, its open-source agent harness, is now a built-in provider in T3 Code. Vercel says it runs on web, desktop and mobile and works with a user's existing models and skills. Vercel published a changelog entry, and its post had 39 points on X by October 28.Microsoft brings the Copilot harness to Foundry Agent Service
Microsoft introduced the Microsoft Copilot harness in Foundry Agent Service. Microsoft says GitHub's engineering team built a multi-model harness that it has now implemented across Microsoft products, and that the Foundry version lets developers build agents without rebuilding the execution layer. According to Microsoft, the harness handles orchestration, tools, context and controls, so developers can focus on their agent's skills and outcomes. Microsoft published a demo video and a Foundry blog post.
GitHub Copilot in VS Code adds cloud sessions that follow you across clients, an Agent Host and experimental computer use
GitHub's VS Code update, announced at Universe, focuses on where agents run and how developers stay involved.GitHub Copilot Fusion, which routes and orchestrates models for each task, is generally available
GitHub says Copilot Fusion is generally available. Instead of choosing a model, a user picks Copilot Fusion in the model picker and an effort level (Standard, High or Max); Copilot decides which models to use and how they work together. It combines the multi-model orchestration of the earlier Project HydraFusion preview with Copilot's auto model selection. Workflows include a single model, a cascade (an efficient model drafts and a quality check decides whether to escalate) and critique (one model drafts, another reviews, the first revises).
GitHub Copilot app adds Projects and computer use in public preview
GitHub says the GitHub Copilot app now has Projects, in public preview: one place for work that spans several repositories or sessions, or that has no repository at all. A user describes a goal and adds context such as repositories, design mocks, Slack threads or meeting notes, picks models and tools, and Copilot runs as many agents as needed and returns results such as pull requests for review. Project automations cover repeatable tasks, for example creating a promo video whenever a feature lands. Context from connected tools such as Slack and Figma can be used to research a request or draft a specification.GitHub Copilot app gets a security extension for code scanning alerts, in public preview
GitHub says a new security extension lets users work through a repository's code scanning alerts inside the GitHub Copilot app. Alerts are listed by severity with search and filters; a user can open an alert, see linked pull requests, dismiss it (sent for approval where delegated dismissal is used), or ask Copilot to explain or fix one or several. Fixes are made only in the local working copy; nothing is committed or pushed. The /security-review command checks current local changes and shows findings next to existing alerts.GitHub Copilot Memory is generally available, with "dreaming" and cross-repository code search
GitHub introduces "GitHub intelligence", a context layer for Copilot agents built from GitHub data, with three parts:GitHub Copilot Automations can now be shared across a team
GitHub says shared Copilot Automations are rolling out. Until now automations were personal, run with the creator's identity and AI credits. A shared automation belongs to the repository: anyone with read access can see it and anyone with write access can edit it, with change history. Its AI credits are billed to the organization or user that owns the repository, its actions are attributed to Copilot, and it keeps running if its creator leaves the team. Example automations: reproduce a bug when an issue is opened, capture screenshots when a pull request is ready, investigate production incidents when a monitoring tool detects them.GitHub Copilot repository plugin defaults and enterprise plugin controls are generally available
GitHub says repository plugin defaults and enterprise-managed plugin controls are generally available in VS Code, Copilot CLI and the GitHub Copilot app.GitHub Agentic Workflows, an open source project from GitHub Next, is generally available
GitHub says Agentic Workflows is generally available on github.com for Free, Team and Enterprise. Users write automations in natural-language Markdown files, which compile to standard GitHub Actions YAML, so they reuse existing runner groups and policies. Tasks include issue triage, CI failure analysis, documentation updates and test writing, using GitHub Copilot or a bring-your-own coding agent; the default Copilot CLI engine needs a Copilot license.GitHub Copilot in JetBrains gets Fusion, generally available local sandboxes and next edit suggestions
GitHub's JetBrains update brings Copilot Fusion to JetBrains IDEs. Fusion replaces the current Auto model option and is rolling out gradually to all Copilot users; GitHub says its use is discounted 30% for the first 30 days.Vercel's fx v0.0.15 cuts memory use by 70% in long sessions and adds /fork
Vercel released v0.0.15 of its fx CLI. The figures are Vercel's own.Mistral opens a Responses API in public preview and serves DeepSeek V4.1 Flash and GLM 5.3 Flash
Mistral announced on October 28 that its Responses API, made for building agents on Mistral, is in public preview. According to Mistral, each response returns the model's answer, its reasoning and its tool calls as separate items, so an app can handle every step more easily than with Chat Completions. It works with all chat models hosted by Mistral, including Large 4. The documentation is on Mistral's Studio docs site.
OpenAI and ServiceNow expand their partnership to connect ServiceNow workflows with ChatGPT and Codex
OpenAI says it is expanding the partnership it announced with ServiceNow in January, to bring OpenAI models and agents into ServiceNow products and ServiceNow workflows into ChatGPT and Codex.T3 Code adds a single MCP server to control a whole fleet of machines
Theo announced on X on October 28 that T3 Code can now control a whole fleet through a single MCP server. In a demo video he starts 8 threads on different machines from his phone through Grok Bot. He clarified that this runs over the internet, not through computer use: with his laptop turned off, he can still control his other machines from Grok Bot on his phone. The post had about 950 likes and 147 replies by October 28.Zhipu opens Z Managed Agents, a cloud agent runtime for GLM, to all users
A post on X by the account leo114119 says Zhipu's Z Managed Agents is now open to everyone. The post describes it as a cloud-hosted runtime for agents that spares developers from building their own agent runtime, execution environment and task orchestration.