Search “Agents”

1Password Launches 1Password for Agents, Letting AI Agents Complete Logins Without Seeing Passwords

1Password has launched 1Password for Agents, connecting its password vault to AI agents: when an agent reaches a login form, it authenticates to 1Password using a digital key; after the user confirms which credentials are needed and what they are for, 1Password fills in the form directly—credentials are never handed to the agent and are removed once login succeeds, so the agent cannot read or reuse them. The integration currently supports only Meta's Muse and Grok Bot, with OpenAI Dots and Anthropic Claude Duet to follow.

Google opens the infrastructure behind Gemini agents as a cloud service, Agent Compute Engine

Google announced that it will open a preview of Agent Compute Engine to Google Cloud users later this year, providing each AI agent with a hardware-isolated microVM that balances container startup speed with sandbox security. The service was originally internal infrastructure for Gemini, Antigravity, and Google's model training pipelines; when an agent is idle, it checkpoints the microVM to free up resources. Google is also offering an open-source runtime, Agent Substrate, which can run on any Kubernetes cluster; the company says that in a benchmark of 200,000 agents it achieved 500 activations per second, with suspend and resume taking under half a second.

Meshy Launches Official Plugin for Hermes Agent

Meshy has announced that it is now available in Hermes. According to a quote post from Nous Research, Hermes Agent now has an official Meshy plugin that generates 3D models through Meshy's MCP server; in Hermes Desktop, models can be rendered directly in chat, rigged character animations can be played inline, and models and prompts can be saved to a gallery.

How Playwright 1.60 Went from a Testing Tool to the Browser Operation Layer for AI Agents

The article argues that Playwright 1.60 has shifted from a testing framework to a runtime for Agents to operate browsers, that Test Agents (planner, generator, healer) have become production-ready, that the MCP server has matured further, and that ARIA snapshots now include bounding boxes. The author breaks this change down into four layers: perception, execution, diagnosis, and roles. ARIA snapshots provide structured page information; browser.bind() lets exploration, the test runner, and MCP share the same browser instance; traces are folded into HAR network capture; and planner, generator, and healer form a closed loop from plan to heal. The article also stresses that the healer does not touch assertions, and that fixes should be submitted as PR diffs for human review rather than merged automatically.

iX Workshop: Agentic Engineering with GitHub Copilot

iX launches the online workshop "Agentic Engineering statt Vibe Coding", explaining how to integrate coding agents into a controllable development process using GitHub Copilot, Spec-Driven Development and TDD. The course covers requirement specifications, acceptance criteria and technical planning, demonstrates agents executing tasks and validating results with tests, and compares the workload with and without agent support. The workshop is led by Microsoft MVP Gregor Biswanger, limited to 16 participants, from 09:00 to 17:00.

Cyware Launches Last Mile AI Agents to Turn Shared Threat Intelligence into Daily Priorities

On October 22, Cyware released Last Mile Agents, aimed at small organizations in threat-sharing networks that lack dedicated intelligence teams, evaluating received security advisories by industry, technology, and operating environment. The first releases are Vulnerability Watch and Threat Briefing, supported by a Priority Intelligence Requirements agent; the former generates a daily priority list under "act," "attend," and "track," while the latter generates a daily ranked briefing; automated execution of remediation is on the subsequent roadmap. The announcement did not provide performance benchmarks, pricing, or customer-tested results.

Aerial uses an Agent for automated deployment, replacing the default smart TV launcher without root

Aerial is a tool that can replace the default launcher on Chinese smart TVs without root. It uses a "launcher guard" to overlay the original launcher when it is woken up, and leverages Android notification listening to revive within 0.5 seconds after standby cleanup, so home-screen ads disappear, though boot ads still cannot be blocked. During deployment, connect the TV and computer to the same WiFi, enable ADB and find the TV's IP, then copy the prompt from the Aerial official website and hand it to Agents such as Doubao and WorkBuddy to execute; the user only needs to allow USB debugging in the TV pop-up, and the whole process takes about five minutes. The tool—from code and APK to the official website—was produced entirely through AI Vibe Coding, gained over 200 stars within one day of launching on GitHub, and subsequently added features such as icon hiding and adding signal input sources to the home page, along with adaptation for multiple TV models.

Browser Use Roundup of the Best Browser Agent MCP Servers of 2026

Browser Use has released its 2026 roundup of browser Agent MCP servers, dividing such servers into two modes—driver and delegate—and evaluating four of them one by one: Browser Use Cloud MCP, Browser Use open-source MCP, Hyperbrowser MCP, and agent-browser chat. In a comparative evaluation across 100 BU Bench V2.1 tasks, the author found that the four driver-type tools consumed a median of 2.1M to 6.7M input tokens per task, all of which count toward the coding Agent's context. The article also provides each server's installation commands, model selection, billing methods, and known limitations, and explains that long tasks, parallel tasks, and structured-output scenarios are better suited to the delegate mode.

Google Open-Sources AQuA Quality Agent: What Should Test Teams Hand Over to It

On October 8, the Google Developers Blog published a reference implementation of AQuA (Ambient Quality Agent), which runs a quality agent alongside production agents, responsible for clustering failed conversations, reviewing issues, and anchoring diagnoses to deployed source code snapshots. The article argues that it will not make testing disappear, but rather hands over to test teams the tasks of defining anomalies, calibrating evidence, and bearing release responsibility, and presents three validation methods: injecting known faults, blind review controls, and root-cause replay.

Google Open-Sources AQuA Quality Agent: What Should Test Teams Hand Over to It?

On October 8, the Google Developers Blog published the AQuA (Ambient Quality Agent) reference implementation, which places a quality agent alongside production agents, responsible for clustering failed conversations, reviewing issues, and anchoring diagnoses to deployed source code snapshots. The article argues that this type of outer-loop quality inspection may still share blind spots with production agents in models, logs, and business definitions, and that test teams need to shift toward designing observability surfaces, calibrating anomaly labels, maintaining evidence chains, and defining escalation rules. The article also presents three validation methods—injecting known faults, blind-review controls, and root-cause replay—as well as a phased adoption path for small teams.

Stanford CS 329Z Lecture 9 Outlines the Selection of Agent System Optimization Methods

Lecture 9 of Stanford's CS 329Z course presents a selection diagram for Agent system optimization: the prerequisite for optimization is having data, a system, and evaluation metrics all in place; generally, context optimization (prompts and few-shots) is done first because it is cheap, fast, and sample-efficient; when context optimization is insufficient, or when you want to distill capabilities into a small model, or when prompts grow longer and longer, then you move on to changing the weights. The lecture notes also review reflective prompt optimization methods such as OPRO, MIPROv2, TextGrad, and GEPA; weight optimization paths such as SFT, RL (PPO vs. GRPO), and on-policy distillation; and approaches that let agents self-optimize, such as the Darwin Gödel Machine, Meta-Harness, and karpathy's autoresearch. They also point out that agents are already decent optimizers, making evaluation metrics the link that most needs to be controlled.

NVIDIA Releases OpenShell Reference Architecture, Governing Agent Security with a Three-Layer Approach

NVIDIA has released the OpenShell reference architecture, arguing that Agents cannot govern themselves and that security and governance should be placed at the infrastructure layer, which Agents cannot reach. The approach has three layers: the open-source runtime OpenShell defines permission boundaries through Gateway, Supervisor, Sandbox, and a formally verified Policy Prover; Vera CPU provides 88 Olympus cores and 1.2 TB/s memory bandwidth, with CoreWeave measuring sandbox startup more than 3x faster and Agent tasks on Terminal Bench 1.8x faster; BlueField-4 DPU plus Sentry provides optional hardware-level monitoring that can isolate violating Agents within a few milliseconds.

Netflix Research Scientist Shares a Roadmap for Evaluating Agents on Subjective Tasks

Cameron R. Wolfe, a senior research scientist at Netflix, has published a roadmap for evaluating subjective tasks, proposing a three-step process: human evaluation, offline automated evaluation, and online monitoring. The article presents statistical tools such as power analysis, confidence intervals, and repeated sampling, along with five criteria that evaluation sets must meet: diversity, realism, sufficient difficulty, discriminative power, and accuracy. The overall process is: offline evaluation screens ideas; promising versions are handed to human evaluation; after A/B testing, they are fully launched and continuously monitored, with the results feeding back into the evaluation set.

Librarian Agents paper: letting agents organize long contexts themselves to improve retrieval performance

Chelsea Finn retweeted a new paper, Librarian Agents, by Anikait Singh and others, which studies whether agents can adapt to large amounts of unstructured context in specialized domains by organizing their own long-context workspaces and anticipating queries. She notes that agentic search can scale to information of any size but tends to miss relevant content, while optimizing information organization for downstream use can also bring 30+% improvements in OOD domains. The tweet includes links to the paper and code.

AAIF: An Adaptive Agent Framework Based on Strands Agents

AWS built the Adaptive Agent Intelligence Framework (AAIF) with Strands Agents, using llama3.2:1b as the orchestration layer to classify requests: routine tasks are handled by a local SLM via Ollama, while complex analysis is routed through boto3 to frontier models on Amazon Bedrock. This architecture can run on edge devices such as a Raspberry Pi with 8 GB of memory, or be deployed fully in the cloud or in a hybrid setup, thereby reducing API costs while maintaining low local latency.

How Cadence Used ElevenAgents to Cut Time to Reach High-Risk Patients from 1 Hour 48 Minutes to 3.5 Minutes

Cadence used ElevenLabs' ElevenAgents to build a remote monitoring alert agent, reducing the median time to reach patients after abnormal vital signs appeared from 1 hour 48 minutes to 3.5 minutes, raising the answer rate for critical calls from 41% to 60%, and achieving a 47% higher rate of patients answering agent calls than manual calls. The agent calls patients when readings are abnormal, performs symptom screening, and either transfers them to a nurse or closes the alert; by design, it treats "uncertain" as not ruled out, prefers to escalate to a nurse when in doubt, and requires explicit negative answers to all symptom questions before closing an alert.

Google Agents CLI Launches Extension System: From ADK to LangChain, All Can Plug In

Starting with version 1.5.0, Google Cloud's Agents CLI introduces an Extension System that allows developers to override or add commands without forking the tool; version 1.8.0 debuts the first LangChain extension, replacing ADK-bound commands with equivalent LangChain implementations. Extensions are declared via agents-cli-extension.yaml in a Git repository and can override built-in commands such as playground, deploy, and eval generate, or add new commands; enterprise platform teams can use this to enforce policies such as SBOM scanning in deployment pipelines.

Nvidia launches Open Agent Safety Platform, using BlueField-4 DPU to monitor agents

At GTC Berlin, Nvidia launched the Open Agent Safety Platform, adding hardware-level controls around the open-source runtime OpenShell; a BlueField-4 DPU works with Sentry software to independently monitor agent behavior and, when it crosses boundaries, isolate and stop it within milliseconds. OpenShell supports third-party agent frameworks such as Anthropic's Claude Code and OpenAI's Codex, and Nvidia says more than 100 organizations already use the platform's technology. Nvidia also introduced the Agent Knowledge Foundation and the Auto-Ontology library, designed to let agents access structured and unstructured data without enterprise data leaving local premises, and expanded its Nemotron 3.5 Super VL, Nemotron 3.5 Bolt 74B-A9B and Kumo series models.

English translation is being prepared.

After an AI Agent fails acceptance: how do review comments feed into the next round of guidance?

In the default path of AI Tutor Engine, when a student already has a guidance session and continues to ask questions on the same task after submitting for acceptance, the detailed review comments are not automatically loaded into the next model request; the model can only see a status and score summary. The author used unique markers to inspect requests, verified two reachable paths—initial full-history initialization and the student actively pasting the comments—and proposed a feedback contract that selects feedback in submission and review order, limits the context budget, and updates outdated comments. This work did not complete a production fix, nor did it evaluate the guidance effect of a real model.