AI Agent News: Releases, Launches and Usage Jumps, Live · page 24
-
Agent Development Kit (ADK) 2.13.0 released
Stability: Access 12 mature experimental features now promoted to stable status. See the unit guide for more details. (ea298d4); Classifiers: Route workflows with TypeSafe's Jev models via the new TypesafeClassifier and ClassifierNode. Jev returns calibrated…
-
How Bayer Crop Science built an agentic decision engine on Amazon Bedrock AgentCore with AWS Unified Operations
Bayer Crop Science built a multi-agent platform on Amazon Bedrock AgentCore enabling natural language queries on analytics data for breeders and researchers.
-
Langfuse 4.69.0 released
feat(ai-gateway): show per-tool setup instructions after creating a gateway key; feat(topics): write rules as target trace transcript; feat(blob-storage): add an outbound host allowlist for blob storage endpoints
-
AI Agent Memory: How LLM Agents Manage State in Production
Article on AI agent memory architectures and how to build workflows using n8n and Cognee that retain context across sessions.
-
Context7 0.6.0 released
158ad26: search() JSON results now type the rank field on each code and info snippet: the snippet's 1-based place in the reranked order shared by both lists. (+1 more release that day)
-
Zapier Agents: Connection errors affecting Salesforce Pardot Zaps
Major impact; resolved after 84 h 11 min.
-
LiteLLM: Claude Code can leak a GitHub CLI token into logged responses
LiteLLM's incident post on the 1.107.0 release says Claude Code can, in specific situations, put a user's GitHub CLI OAuth token into a response body, so gateways that log responses may store it. LiteLLM believes the attacker took such tokens from its own logs (as reported). No Anthropic statement found yet. Treat proxy logs of Claude Code traffic as secrets.
-
CopilotKit pricing changed: Free + $39/mo → Free + $49/mo
The SDK is free (MIT). The hosted CopilotKit Intelligence platform has a free tier with 1,000 interactions/month. Paid plans use pay-as-you-go pricing at $49/month (includes 10,000 interactions), with graduated rates for additional interactions beyond that.
-
Zapier Agents pricing changed: Free + $33.33/mo → Free + $19.99/mo
Zapier now uses unified task-based pricing across all products including Agents. Free plan: 100 tasks/month. Professional: from $19.99/month. Team: from $69/month (25 users). Enterprise: custom pricing. All plans include Zaps, Tables, Forms, and access to Agen
-
LiteLLM 1.107.0 carried a credential stealer; Railway and Zeabur report leaks of customer secrets
LiteLLM says its release pipeline was compromised and litellm 1.107.0 on PyPI and Docker carried credential-stealing malware; it advises pinning 1.106.2 and rotating keys, including provider keys stored in the proxy. Railway and Zeabur, as reported, had customer environment variables read by attackers seeking AI service keys. The post's wider list of 27 incidents was not checked by us.
-
Agent-Reach 1.6.0 released
从旧版升级:把"帮我更新 Agent Reach"那句话发给你的 AI 就行。你的密钥和设置都保留,旧版程序会自动删掉。; 加一个平台只要说一句,需要的软件都会自动装好,不需要管理员密码;你只需要登录或者填一个 Key。; 小红书、推特这类要登录的平台,改用 Agent Reach 自己的专用浏览器,不弹窗、不占用你的浏览器;多个 AI 可以同时查。
-
Dify 1.18.0 released
Published triggers: the draft-sync fix does not repair subscriptions already removed before upgrading. Rebind missing subscriptions and republish affected workflows as needed.; Document extraction: re-import/re-index affected content to apply extraction…
-
OpenAI Codex 0.164.0 released
Save reusable prompts and slash commands in ten numbered /presets slots, with a keyboard menu for quick access.; Set model, reasoning effort, speed, and permission profiles directly with inline /model and /permissions arguments.; Preview viewed and generated…
-
Pydantic AI 2.60.0 released
Raise model error categories like ModelRateLimitError from embedding models; Add usage_limits, deadline, run_id and conversation_id to ModelSelectionContext; Default BedrockConverseModel's maxTokens to Claude's maximum output, streaming such requests behind…
-
GitHub Copilot in Visual Studio — September update
GitHub Copilot in Visual Studio added features including pull request exploration in chat, team model selection, and vulnerable package fixes.
-
Anima helps clinics cut call waits by 75% with GPT-6 and GPT-Live
Anima reduced clinic call wait times by 75% using OpenAI's GPT-6 and GPT-Live models to expand care team capacity.
-
Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed are available now in AI Gateway
Kimi K3.1 Pro and K3.1 Pro Highspeed from Moonshot AI are now available on Vercel's AI Gateway, with improvements in coding, 3D modeling, financial analysis, and agentic tasks.
-
AgentScope downloads up 27% in a week
52.1K PyPI downloads in the 7 days to Oct 23, against 41.1K the week before.
-
Graphite downloads up 42% in a week
37.4K npm downloads in the 7 days to Oct 23, against 26.3K the week before.
-
Haystack downloads up 31% in a week
152K PyPI downloads in the 7 days to Oct 23, against 116K the week before.
-
PI Mcp Extension downloads up 49% in a week
133K npm downloads in the 7 days to Oct 23, against 88.9K the week before.
-
OmniRoute passes 80K GitHub stars
-
smolagents passes 30K GitHub stars
-
Anatomy gained 5,614 GitHub stars in a week
Now at 19.8K stars.
-
CLI-Anything gained 3,101 GitHub stars in a week
Now at 56.6K stars, against about 1,052 a week over the month before.
-
Ego Lite gained 1,569 GitHub stars in a week
Now at 19K stars, against about 349 a week over the month before.
-
Ruflo gained 3,206 GitHub stars in a week
Now at 78.7K stars, against about 800 a week over the month before.
-
Solidcraft gained 6,606 GitHub stars in a week
Now at 9,737 stars.
-
Xerj gained 7,169 GitHub stars in a week
Now at 14K stars, against about 1,260 a week over the month before.
-
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
ThinkingBox enables post-training of agents using database grading results converted into reinforcement learning rewards.
-
Custom skills, agents, and MCP servers in usage metrics API
GitHub Copilot usage metrics API now displays names of custom skills, agents, MCP servers, slash commands, and plugins instead of grouping them.
-
Deep Agents 0.1.0 released
deepagents-talon now requires Python 3.14 (>=3.14,<3.15).; Added /effort controls for reasoning.; Added the NVIDIA OpenShell sandbox provider and an openshell extra for Talon. (+3 more releases that day)
-
Donating Agent Skills to the Agentic AI Foundation
Anthropic is donating Agent Skills standard to the Agentic AI Foundation under the Linux Foundation, joining Block and OpenAI as co-founders.
-
Stream low-latency speech from AI agents using Kokoro TTS on Amazon SageMaker AI
AWS shows how to stream low-latency speech from AI agents using Kokoro TTS on Amazon SageMaker AI by synthesizing each sentence as produced.
-
OpenShell – A safe, private runtime for fleets of autonomous AI agents
OpenShell is a safe, private runtime for fleets of autonomous AI agents.
-
Hallucination Detection Methods for AI Agent Workflows
n8n discusses hallucination detection methods for AI agent workflows to catch false information before it cascades to downstream steps.
-
Lovable: Some users may see errors when setting up a database or backend for a new project, or when remixing a project.
Major impact; resolved after 36 min.
-
Junie 3897.1 released
(+9 more releases that day)
-
How Rescale built an AI triage agent with Strands Agents and AI-DLC
Rescale built an AI test-triage agent using Strands Agents and Amazon Bedrock with AWS's AI-DLC framework.
-
E2B 2.59.0 released
e5a9fa6: Add dnsServers to the network options (dns_servers in Python): the DNS servers a sandbox can reach when allowOut contains a domain.; 5a714f0: Deprecate the apiUrl / api_url and sandboxUrl / sandbox_url connection options and their E2B_API_URL and…
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.