AWS previews Strands Box for controlling AI agent actions
Strands Box combines local isolation with rules based on earlier actions. Its macOS preview leaves direct file grants outside policy history.
AI Agents
AI agents bring language models into workflows that involve planning, using tools and carrying out tasks across software. ByteForward follows coding agents, browser automation, application integrations and assistants that work through several steps toward a goal. Coverage focuses on the permissions, supported actions and human controls that shape what an agent can actually do. Reliability also depends on the tools, task environment and checks around the model. Costs, execution limits and recovery from mistakes are practical considerations for anyone adopting automated AI workflows. Our AI tools reporting covers the applications and developer platforms that make these workflows available. Explore AI safety and security for reporting on prompt injection, containment and other risks connected to agent access. Use the articles below to understand new capabilities alongside the oversight they need.
LangChain adds pinned skills and scheduling to Deep Agents
The October 7 updates add skill pinning and a managed scheduling beta, with version requirements and US cloud limits.
OpenAI releases desktop fixes for ChatGPT Work thread failures
Updated desktop versions restore thread creation across Linux macOS and Windows.
Selective agent audits use fewer tokens but miss more unsafe runs
A new preprint measures the cost and detection tradeoff in agent monitoring.
Procuros raises €20 million for AI trade infrastructure
XAnge leads a Series A for supply chain data connections that support automated trade.
Vocca raises $20 million to expand healthcare AI agents
Norrsken leads a Series A for AI agents that handle medical appointments and routine patient calls.
GitHub starts fixing missing Copilot agent usage data
GitHub says missing agent usage data cannot be recovered and billing was unchanged.
Jump Trading describes longer AI research with GPT 6 Astra
OpenAI publishes a customer account of longer quantitative research tasks with continuing human review.
MemPilot research makes AI agent memory spending adjustable
A new preprint tests selective memory processing across five benchmarks, with latency and privacy limits still important.
OpenAI tests Astra on Ironclad contracting tasks
An 11 task research evaluation compares Astra and Sol on contracting workflows.
Atlassian expands OpenAI partnership for Rovo agents
The October 6 agreement broadens model access while deeper Jira integrations remain under exploration.
Hack The Box opens AI Range to enterprise security teams
The edition tests agents against cybersecurity job roles.
Google proposes contextual privacy checks for AI agents
A new workshop report examines how agents should handle changing tasks and data sharing.
Claude Code update repairs permission checks and scheduled tasks
Claude Code 2.1.290 fixes permission enforcement and task recovery with a compaction limit.
Instinct opens group chats to early access users
A shared agent can coordinate plans while personal agents check permissions before sharing information.
TikTok unveils AI shopping agents and direct brand checkout
Shopping Assistant, Buy Direct and Lead Agent are available through eligibility based tests arranged with TikTok sales partners.
Dubai launches agentic AI accelerator for government services
The first phase will assess more than 300 potential uses across 33 government entities.
Logistics Reply adds five governed warehouse AI agents
Five warehouse agents became available on October 5, alongside a framework for setting authority by task, operational context and organizational readiness.
Antigravity CLI removes Windows sandbox admin requirement
Version 1.2.17 makes sandboxed commands easier to run
Replit’s coding agents solved more tasks with flexible delegation
More flexible delegation raised benchmark scores and the cost per task.
Oracle Fusion Claw adds governed execution for business agents
Oracle says 25 applications use Fusion Claw to combine AI reasoning with controlled execution
Tuskira opens its AI Agent Gateway
Tuskira publishes an alpha gateway for AI agents
Gemini Enterprise adds live queries across cloud databases
Google previews database connectors with broad user access and separate controls for stored results and write actions
Meta adds business connectors to Muse
Muse for Small Business connects work apps and business accounts while permissions and data settings need careful review