Topic

AI agents

Autonomous agents, orchestration, safeguards and new ways of working.

AI agents move from answering to acting. This topic follows their practical capabilities, integration into teams and the controls required when a system can use tools or change data.

CybersecuritySeptember 26, 20267 min read

GitHub automates SSO authorization for PATs and SSH keys

Enterprise Cloud administrators can authorize a classic token or SSH key across 50 organizations through a GitHub App.

GitHub EnterpriseSSOPAT

AISeptember 17, 20267 min read

Meta launches Muse, an AI agent that acts on the web

Muse can send email, complete forms and prepare travel, with approval before sensitive actions.

MetaMuseAI agent

AISeptember 17, 20268 min read

ToolGrad builds more reliable tool-use data for AI agents

Google Research proposes generating and correcting tool-use tasks with textual feedback, reporting a 99.8% validation rate.

ToolGradAI agentsGoogle Research

CybersecuritySeptember 12, 20267 min read

Docker argues for secure-by-default AI agents: controls that matter

Docker outlines an approach built on verified images, isolation, least privilege and end-to-end traceability for the agent software supply chain.

DockerAI agentssupply chain

DevelopmentSeptember 03, 20267 min read

Docker Sandboxes isolates AI agents in GitHub Actions with a microVM

Agents can run Docker and fix code inside a disposable microVM while network and GitHub outputs remain constrained.

Docker SandboxesGitHub ActionsAI agents

ComputingAugust 17, 202610 min read

Google opens its lakehouse to AWS and Azure data for AI agents

BigQuery, Apache Iceberg and Knowledge Catalog aim to unify multicloud data without moving everything: the promise, architecture and limits.

Google CloudBigQueryApache Iceberg

CybersecurityAugust 13, 20269 min read

CodeMender lets a Google AI agent find, verify and patch code vulnerabilities

CodeMender scans code, validates flaws with exploits and proposes tested patches. Here is what Google promises and which safeguards teams still need.

Google CloudCodeMenderapplication security

DevelopmentAugust 10, 202613 min read

WebMCP arrives in Chrome 149: preparing websites for agents securely

WebMCP lets websites expose structured actions to AI agents. API design, the Chrome 149 trial, use cases, security risks and a practical test plan.

WebMCPChrome 149AI agents

CybersecurityAugust 03, 20269 min read

Microsoft launches Project Perception, using AI agents to prioritise flaws

Project Perception enters preview with an initial mission: connect security signals, identify attack paths and accelerate software vulnerability remediation.

microsoftcybersecurityai agents

AIJuly 31, 202611 min read

Generative AI for business: use cases, cost, security and governance

A practical guide to selecting an AI use case, evaluating an assistant, protecting data and measuring value before scaling.

aibusinessgovernance

CybersecurityJuly 30, 20268 min read

OpenAI, Hugging Face and the runaway agent: cybersecurity is changing pace

An incident involving a research agent shows why autonomous systems need real incident procedures, not only prompt-level guardrails.

agentsopenaiincident

CybersecurityJuly 29, 20268 min read

Claude Cowork and SharedRoot: a VM is not enough to secure an AI agent

Accomplish AI's SharedRoot chain shows why local agents must be treated as untrusted execution environments, even when they run inside a virtual machine.

claudeai-agentssandbox