AI Agent News: Releases, Launches and Usage Jumps, Live · page 12
-
Claude Mem 15.11.0 released
Track the nine actual setup screens: Scan, Reveal, Choose memory, Review savings, Checkout, Finish setup, Import history, Launch, and Workspace.; Count each rendered screen visit separately. Returning to a screen creates a new visit; failed hand-offs do not…
-
Copilot impact dashboard adds chat and VS Code agent usage
GitHub's Copilot impact dashboard now tracks usage across nine Copilot experiences, including Chat and VS Code Agents.
-
Beyond code generation: A developer’s guide to agentic engineering
GitHub published a guide on agentic engineering with Copilot for planning, code review, and production issue investigation.
-
Junie 4040.1 released
(+8 more releases that day)
-
GitHub Copilot weekly releases — November 2
GitHub Copilot now available in JetBrains Air and Xcode, with updates to pull request management, settings, and agent sessions.
-
AI Gateway adds DeepSeek Search and Fetch tools
Vercel's AI Gateway now includes DeepSeek Search and Fetch tools for retrieving and reading web content across any model supporting tool calling.
-
Chat SDK agents can now stream generative UI with json-render
Chat SDK agents can now stream generative UI using json-render, which converts JSON specs to platform-native UI formats like Slack Block Kit and Teams Adaptive Cards.
-
Mintlify Assistant: Docs search unavailable
Critical impact; resolved after 17 min.
-
Design the Muse App, Just for You
Today, we're introducing spaces, an interactive way to build a more customized experience in the Muse app. The post Design the Muse App, Just for You appeared first on Meta Newsroom .
-
Perplexity launches AI agents for work
Perplexity launched AI agents for work.
-
Taxonomy of agentic primitives in AWS Quick
AWS shares a taxonomy of agentic primitives from building Amazon Quick, organizing agents around triggers, capabilities, learned context, and decision-making.
-
Serverless MCP server for AI agent data lake access on Amazon ECS Fargate
AWS published a guide on building a serverless MCP server on Amazon ECS Fargate that enables AI agents to securely access enterprise data lakes with JSON-based dataset configuration.
-
Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 2
AWS describes scaling governed AI agent access to enterprise scale using Amazon Bedrock AgentCore Gateway with Okta integration, Cedar policies, and observability across multiple agents.
-
Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 1
AWS published guidance on building a task agent on Amazon Bedrock AgentCore that enforces user-aware access control using Okta Cross App Access and AgentCore Identity.
-
Taxbit Assist: A grounded, role-aware AI agent on Amazon Bedrock AgentCore
Taxbit built Taxbit Assist, a cited permission-aware AI agent on Amazon Bedrock AgentCore with role-based safety controls and quality measurement on live production traffic.
-
Extend Claude Code to remote sessions on Amazon Bedrock AgentCore with security controls
AWS describes running Claude Code as a remote agent on Amazon Bedrock AgentCore with isolated sessions and pull request output for code changes.
-
CopilotKit 1.85.0 released
This release introduces native Generative UI for Angular, adds analytics context forwarding for Intelligence runs, and fixes several reliability issues around Intelligence run stops and A2UI…
-
Fin: Elevated errors and latencies in the European hosting region.
Major impact; resolved after 49 min.
-
[AINews] Inkling 2 Small: 17A-594B trained in 68 hours on 12T tokens, beating Muse Spark
November 5, 2026 — Inside Inkling 2 Small’s training run and METR’s study of AI research at Anthropic.
-
Oh My Pi 18.17.0 released
Added Agent.queuedMessageGate to filter queued messages before delivery and observe claimed messages returned undelivered.; Added wireToolsFingerprint(), returning a hash of the tool specs sent with a request.; Added opt-in standard Responses WebSockets for…
-
OpenAI Codex 0.167.0 released
Open the agent command center with codex agents --no-daemon. The command center and session picker now mark unread results.; Switch between connected backends from the terminal session picker and keep your drafts when switching.; Preview local image…
-
Measuring how AI agents handle everyday wet-lab science
OpenAI introduced BioBasicsBench, a 100-task benchmark for measuring how well AI agents interpret raw experimental data in wet-lab science.
-
Voice Chat with Agents
Cursor added voice chat capability for agents, enabling realtime voice calls on desktop, web, and iOS.
-
Browser Harness downloads down 30% in a week
2.3M PyPI downloads in the 7 days to Nov 6, against 3.2M the week before.
-
Inkeep downloads down 35% in a week
144K npm downloads in the 7 days to Nov 6, against 220K the week before.
-
kapa.ai downloads down 35% in a week
31K npm downloads in the 7 days to Nov 6, against 47.5K the week before.
-
Amp downloads up 32% in a week
50.6K npm downloads in the 7 days to Nov 6, against 38.4K the week before.
-
eve downloads up 29% in a week
2.4M npm downloads in the 7 days to Nov 6, against 1.9M the week before.
-
Godot MCP downloads up 42% in a week
40.1K npm downloads in the 7 days to Nov 6, against 28.1K the week before.
-
Loki Mode downloads up 52% in a week
20.7K npm downloads in the 7 days to Nov 6, against 13.6K the week before.
-
e2e downloads up 28% in a week
766K npm downloads in the 7 days to Nov 6, against 601K the week before.
-
Watch passes 20K GitHub stars
-
Dsh Superagent Prototype passes 20K GitHub stars
-
HyperFrames passes 80K GitHub stars
-
Sim passes 30K GitHub stars
-
Anatomy gained 2,545 GitHub stars in a week
Now at 25.6K stars.
-
Day Trader Skill gained 2,123 GitHub stars in a week
Now at 5,812 stars.
-
Learnharness gained 1,523 GitHub stars in a week
Now at 5,841 stars.
-
OpenMausBot gained 6,988 GitHub stars in a week
Now at 15.9K stars, against about 1,238 a week over the month before.
-
PR-Agent gained 1,565 GitHub stars in a week
Now at 15.3K stars, against about 149 a week over the month before.
This list shows notable items; the sections above show everything of their kind, including patch releases of smaller projects and low-impact incidents. Summaries of outside articles are written automatically from the source's own text and may miss nuance, so follow the link before relying on one.