AI Agent News: Releases, Launches and Usage Jumps, Live
A live feed of what is happening with AI agents: new releases, launches from official blogs, download and GitHub star jumps, outages and the most discussed stories, each linked to its source. Updated every 30 minutes.
-
Claude Mem hits 155.3K weekly downloads, 257% above its recent average
Claude Mem, a memory plugin for Claude Code, reached 155.3K npm downloads in the week ending Nov 19, 2026. That is more than double its previous weekly record of 75.3K.
-
Mastra sets a weekly download record, 36% above its 4-week average
Mastra, a TypeScript framework for building agents and AI workflows, reached 4M npm downloads in the week ending Nov 18, 2026. That is its highest weekly total in 75 weeks of data.
-
Fin Voice degraded for 1 h 35 min, 2.1x the typical incident
Fin, an AI customer agent for support chat, email and voice, had a minor Fin Voice incident on Nov 18. It ran from 16:27 UTC to 18:02 UTC, about twice as long as Fin's typical incident.
-
ClawDOS – An AI coding agent for DOS
ClawDOS is an AI coding agent designed for DOS environments.
-
New in AWS Quick: More capable flows, an updated web experience, and agents that work together
AWS Quick adds multi-agent coordination, flow building in chat, web app updates, and new data source integrations including Meta ads and Amplitude.
-
Cut self-hosted LLM cost and latency with the distributed KV caching agent skill
AWS released a distributed KV caching skill for its Agent Toolkit that reduces self-hosted LLM inference costs by up to 40% and latency by up to 80% using ElastiCache.
-
Deep Agents 0.1.0 released
Requires agent-client-protocol>=1.0.0,<2.0.0.; Fixed a crash when running dcode --acp with agent-client-protocol 1.0.; 9dca61f release(deepagents-acp): 0.1.0 (+4 more releases that day)
-
Show HN: OpenComputer – an open source alternative to Devin
An open source project positioned as an alternative to Devin, a coding AI agent.
-
Resolve AI: Degraded performance: data not loading and actions failing in Resolve
Major impact; resolved after 17 min.
-
Cloudflare Agents SDK 0.34.0 released
Pi 1.6 support: agents/harness/pi, agents/models/pi-ai and the Pi tools now run on Pi 1.6, which brings; Tools that wait: a tool can park until a signal arrives or a deadline passes, and the agent can sleep meanwhile. Deliver signals with harness.signal()…
-
CopilotKit 1.90.0 released
This release adds framework-level telemetry to the runtime and sends the CopilotKit version to the Intelligence gateway, giving the gateway the context it needs to tailor behavior to a client's…
-
Claude Code approvals left 31,088 secrets in files on public GitHub, 5,343 still working
When you answer "Yes, and don't ask again" to a shell command, Claude Code saves the whole command as a rule in .claude/settings.local.json, including any API key or password typed into it. GitGuardian scanned public GitHub and Docker Hub for copies of that file.
-
Pi 1.6.0 released
GPT-6.2 Astra and GPT-6.2 Sol: both models are available on the openai, azure, and openai-codex providers, and GPT-6.2 Sol is now the default model for all three. See Providers.; Extension components: render() now returns readonly string[], and Container…
-
NiCE Cognigy: Cognigy.AI – Service Disruption
Critical impact; resolved after 1 h 18 min.
-
Vercel completes ISO/IEC 42001:2023 certification
Vercel completed ISO/IEC 42001:2023 certification for AI management systems, covering its infrastructure and AI products including AI Gateway.
-
Use Kimi Code CLI with models on Amazon Bedrock
Kimi Code CLI v2.7.0 adds native Amazon Bedrock provider support, allowing use of models like Kimi K3.1 Pro via AWS credentials.
-
Hebbia sees improvements in financial research with GPT-6 Astra
Hebbia evaluated GPT-6 Astra for financial research and found improvements in source coverage, citation relevance, and presentation quality.
-
Deep Agents 0.4.0 released
Upgrade to quickjs-rs 0.3.0. JavaScript state saved by 0.3.x cannot be restored; existing threads start with a fresh JavaScript context on their next evaluation.; Keep QuickJS interpreters warm in a process-wide LRU cache.; Add an opt-in, Node-style fs…
-
Enable AI agents to pay for AWS Marketplace x402 products with Amazon Bedrock AgentCore payments
Amazon Bedrock AgentCore payments now supports AWS Marketplace, enabling agents to discover and purchase products, APIs, data, and content with billing through AWS accounts.
-
Agent Development Kit (ADK) 2.15.0 released
Integrations; Verify addresses, reverse-geocode coordinates, and fetch Street View imagery directly from an agent with GoogleMapsToolset (get_street_view_image, validate_address, and reverse_geocode).; Answer open-ended analytics questions without knowing…
-
How the BMW Group governs hundreds of agents across thousands of developers
BMW Group shares governance practices for managing hundreds of AI agents across thousands of developers at enterprise scale while maintaining speed, quality, and compliance.
-
Strands Agents 0.5.0 released
feat(catalog): add strands-acgs (9faf8a49a); feat(tools): allow sorting tool specs (543affec3); feat(todos): add reminder interval (2e33620c0) (+2 more releases that day)
-
Gemini CLI 0.70.0 released
Changelog for v0.69.0-preview.0; Changelog for v0.68.0; chore(release): bump version to 0.70.0-nightly.20261110.g1603e9bab (+1 more release that day)
-
Build trusted autonomous agents for your team with Microsoft Foundry
Microsoft Foundry introduces autonomous agents that work in Microsoft 365 with their own identity, permissions, and accountability structures for enterprise teams.
-
From learning to leaderboard: How Cisco upskilled 1,200 engineers in agentic AI with AWS AI League
Cisco upskilled 1,200 engineers in agentic AI through AWS AI League competitions, with a Cisco engineer winning the 2025 global title and a winning agent built on Amazon Bedrock AgentCore.
-
Protect agent tool calls with Skyflow and Amazon Bedrock AgentCore Gateway
Skyflow integrates with Amazon Bedrock AgentCore Gateway to de-identify sensitive data before agents call tools, then restore it for authorized users.
-
PolyAI Dialog-RSN-1 is now on Amazon SageMaker JumpStart
PolyAI Dialog-RSN-1, a reasoning model for real-time complex dialog, is now available on Amazon SageMaker JumpStart for evaluation and deployment.
-
Launch HN: Onlook (YC W25) – An open-source developer tool for designers
Onlook, a Y Combinator W25 company, launched an open-source developer tool for designers.
-
Skyvern outage lasted 12 h 5 min, 6.2x its typical incident
Skyvern Cloud had a major incident from Nov 17, 12:22 UTC to Nov 18, 00:28 UTC, during which browser sessions and agents failed to start. It is the longest Skyvern incident in the past 12 months.
-
Synthflow ElevenLabs voice incident ran 5 h 32 min, longest in 12 months
Synthflow, a platform for AI voice agents that answer and place phone calls, had a minor incident on Nov 18. Calls and API requests degraded for agents using ElevenLabs voices, and it lasted 13.3x the typical incident.
-
Droid's Gemini models were down for 39 h 51 min, 58.2x its typical incident
Factory's Droid, a software-engineering agent, lost access to its Gemini models from Nov 15, 04:01 UTC until Nov 16, 19:52 UTC. The incident was rated major and is the longest in the past 12 months.
-
Langfuse 4.83.0 released
feat(topics): new Scatter view, remove List; feat(annotation-queues): navigate between items from single queue item view (LFE-18664); feat(native): extract skill requests and share project ID resolution
-
Claude Agent SDK: Elevated errors for Claude Mythos 5.2 and Claude Fable 5.2
Major impact; resolved after 22 min.
-
Claude Code: Elevated errors for Claude Mythos 5.2 and Claude Fable 5.2
Major impact; resolved after 22 min.
-
Claude Cowork: Elevated errors for Claude Mythos 5.2 and Claude Fable 5.2
Major impact; resolved after 22 min.
-
Claude Design: Elevated errors for Claude Mythos 5.2 and Claude Fable 5.2
Major impact; resolved after 22 min.
-
Claude in Chrome: Elevated errors for Claude Mythos 5.2 and Claude Fable 5.2
Major impact; resolved after 22 min.
-
[AINews] Gemini 4 Neon: GDM's Opus/Sol Contender
Google released Gemini 4 Neon, its middle-tier model, which scored 61 on the Intelligence Index. Anthropic priced its IPO at $83.
-
Agents can now find Vercel CLI commands by describing the task
Vercel CLI now supports natural language command search for coding agents via vercel help search, with optional help page inclusion.
-
Explore the Codex Security Plugin
Demo available for Codex Security Plugin showing workflow from setup through results review.
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.