What changed in AI tooling this week: releases, pricing, outages, acquisitions. Every item links to its source.
Calling /create formal verification skill makes the agent pick a high stakes property (races, retries, parsers, unsafe code), choose a tool such as TLA+, CBMC, or Kani, show the check failing on a de…
The open eval (GitHub) checks a from scratch SQLite 3.40.1 engine against 272 tests adapted from SQLite's own suite and scores pass rate alongside API cost; OpenAI says GPT 6 Astra averages 87.4% (CO…
The single file, stdlib only Python tool (macOS only) confines a run to the folders you grant, blocks your home folder, and routes every API call through a loopback proxy that holds the real key, pri…
The YC W24 team rebuilt its LiteLLM based gateway as a tiny event dispatch core, so providers (Mistral's is 22 lines), MCP, realtime voice, fallbacks, and PII redaction are all plugins, and an OpenAI…
Now in free public preview for GitHub Enterprise Cloud, it sits between coding agents or Actions runners and the upstream registry, checking metadata requests and downloads against admin policy (a co…
A dev only SDK, stripped from production builds, instruments your running web, Electron, or Tauri app's DOM, network, routing, console, and state (Redux, Zustand, TanStack Query, Pinia) and exposes i…
Microsoft's security stack now discovers 60 kinds of local agents (35 newly added), registers them under their own Entra Agent ID instead of the user's, and is introducing Dynamic Permissions that st…
The Oct 20 release brings persistent PTYs to Windows x64 and ARM64, an opt in todowrite task list via { "experimental": { "todo": true } } , TUI and plugin commands that can outlive Bun's five minute…
Released Oct 20, a day after 3.0.0, it adds an admin managed enable auto approve = false that hides auto approve agents and ignores yolo , makes vibe p wait for subagents its turn left running and ex…
The internal loop finds where Kiro fails or wastes effort, proposes one prompt or tool change at a time, and keeps it only if a code based gate (including a full benchmark run against a re run contro…
The six models, including GPT 5.4 mini and GPT 5 mini, are gone from Copilot Chat, inline edits, ask and agent modes, and completions; GitHub suggests GPT 6.1 Sol, GPT 6 Luna, Grok 4.7, or Gemini 3.8…
Double tap Control, rest the cursor on the broken UI element, and talk; the free, open source macOS app queues your transcript (speech to text runs on device, no account) with a cropped screenshot th…
The always on agent, now in beta on Pro and Enterprise plans, reads merged diffs, repeated support questions, and flagged Slack threads, drafts the doc change, and opens a pull request that a teammat…
The open source terminal agent's Oct 19 release lets a hosted parent launch snapshot isolated children that can read a frozen copy of the workspace without editing files or running commands, and adds…
Scope (MIT) runs the same task through agents like GitHub Copilot and Claude Code over repeated runs, scores each against a dependency graph of criteria with an LLM judge, and keeps tool calls, logs,…
The MIT repo, built from the setup behind Votito, bundles a sandboxed Claude Code container, a bash command regulator that auto approves or denies with guidance, a lint on edit PostToolUse hook, and…
Fine tuned from Qwen3.8 27B on 127 real engineering sessions, it ties its base at 133/150 on BenchLocal while generating 25% fewer tokens, and its multi token prediction draft head roughly doubles co…
The guided setup learns from your project and recent sessions and now also runs on Bedrock, Vertex, Foundry, and telemetry disabled sessions, while the new /claude api managed agents secure checks a…
The new coding CLI has Agent, Build (parallel tasks gated by tests), and Architect modes; Architect turns an approved blueprint into a plan its Agent OS compiler hands to parallel agents, and Spinoza…
Built on the Pi harness, Fade runs a conversational companion and a Builder side by side so you can plan or line up tasks mid build, and it only marks work "checked" after committing the change and r…