AITOPAI热报

#MCP and tools

1086 stories · 132 in 7 days · 458 in 30 days · Topic watch →
agents 398Claude 94coding assistants 90Claude Code 73Anthropic 53LangChain 52prompt engineering 46ChatGPT 42
Nov 7
ProductsBuildersSelected02:58
LangChain adds browser use and async subagents to deepagents

The pairing of browser control with non-blocking subagents points toward agents that act proactively rather than only responding to prompts, a design shift worth tracking. Developers building agent workflows in Python can try both features now through the deepagents docs and compare the Stagehand approach with Playwright-based alternatives.

#LangChain#deepagents#Stagehand#agents
Official accountLangChain@LangChainAISource
ProductsBuildersOfficial02:45
Chinese contentClaude Code 发布 v2.1.319：MCP 任务后台化与网关多项更新

Anthropic 给 Claude Code 更新了一批实用功能：MCP 任务能直接丢后台跑，权限卡住时自动拒绝不阻塞，还能用 /claude-api latency-optimize 找 API 延迟点，用网关的团队尤其值得升级。

#Claude Code#Anthropic#MCP and tools#coding assistants
Event
Primary sourceClaude Code: GitHub Releases@claude-code-publisher[bot]Source
ProductsBuildersMulti-sourceSelected02:08
Fulcra launches multiplayer mode for cross-platform agent messaging

The design sidesteps the usual trust problem by exposing only per-agent read-only channels instead of shared logins or API keys, with every share revocable at any time. Developers building multi-user agent workflows can wire it up through MCP or the CLI today on the free tier, without waiting on vendors to agree on an interoperability standard.

#Fulcra#Claude#ChatGPT#MCP and tools
Event
rohanpaul_ai@rohanpaul_aiSource
How-toBuildersOfficialSelected01:33
Serverless MCP Server on Amazon ECS Fargate for AI Agent Data Access

The metadata-driven design means teams can onboard new datasets without writing integration code, shifting agent data access from bespoke connectors to configuration. Builders deploying agent frameworks on AWS can use the guide's ECS Fargate pattern as a starting architecture. It contrasts with fully managed options like AWS Bedrock's built-in MCP support by keeping control of the serving layer.

#MCP and tools#Amazon ECS#Fargate#agents
Primary sourceAWS Machine Learning Blog@Aishwarya PawarSource
How-toBuildersOfficialSelected01:32
Okta Cross App Access with Amazon Bedrock AgentCore for agent governance

The pattern shifts agent security from per-app ad hoc controls to centralized, auditable authorization: identity is redeemed once at the gateway, and Cedar decides tool-level access per agent. Platform and security engineers building multi-agent estates can replicate it with the supplied CloudWatch and Okta log correlation for compliance audits. It pairs with Part 1, which covered the single-agent setup.

#Amazon Bedrock#AgentCore#Okta#Cedar
Primary sourceAWS Machine Learning Blog@Hammad AusafSource
Nov 6
ProductsBuildersMulti-sourceSelected03:32
vLLM Ecosystem Powers Kimi K3 and GLM-5.2 on TPUs

TPU inference for Chinese open-weight models signals that serving stacks are diversifying beyond NVIDIA, with vLLM TorchTPU as the concrete path to run models like GLM-5.2 and Qwen3.5 397B on Google silicon. Teams evaluating inference cost and hardware options should follow the SemiAnalysis write-up for benchmark details on the AgentX results.

#vLLM#Kimi K3#Qwen3.5#TPU
Event
Official accountvLLM@vllm_projectSource
How-toBuildersSelected00:13
Bitkub MCP Package Re-Registration Exposes Agent Supply Chain Risk

The deeper point is that package-name continuity proves nothing — identity of the maintainer is what matters, and stale docs can silently redirect installs. Teams wiring agents to financial or exchange APIs should audit publisher identity before every upgrade and enforce limits in execution code rather than prompts. The six-point checklist is directly usable as a review template for any MCP deployment.

#MCP and tools#AI safety#Bitkub#Agent
岚叔@lufzzlizSource