Agents
-
Research
Agent study finds lost authorization constraints, not context compaction, drive control failures
A Tencent Zhuque Lab study finds agent control failures rise when context management drops authorization rules, while constraint-preserving compaction remains safe in its benchmark.
-
Agents
OpenAI turns the Codex agent harness into a managed API
OpenAI launches Agents API in public beta, separating managed Codex orchestration from selectable execution environments and changing the agent infrastructure boundary.
-
Agents
Anthropic’s threat report shows AI moving from cyber assistant to operational loop
Anthropic’s September 2026 report shows Claude used in multi-step malicious workflows, shifting security attention from model outputs to operational loops.
-
Agents
GitHub turns Code Quality backlog fixes into a bulk Copilot agent job
GitHub Code Quality can assign up to 25 findings to Copilot at once, shifting remediation toward batch agent work while preserving pull-request review.
-
Industry
Red Hat AI 3.5 turns AI governance into platform infrastructure, with a support-tier catch
Red Hat AI 3.5 makes EvalHub GA and expands identity, GPU and agent controls, while some observability remains Technology Preview.
-
Agents
Mastra Factory beta makes coding-agent autonomy a stage-by-stage decision
Mastra Factory enters beta with configurable human and agent gates across issue intake, planning, coding and review, plus self-hosting options and caveats.
-
Agents
GitHub makes Copilot agent permissions enforceable above local settings
GitHub adds enterprise-managed deny, ask and allow rules for Copilot agent operations that local auto-approval and saved grants cannot weaken.
-
Agents
Anthropic's fourth cyber incident exposes a gap in the audit itself
Anthropic found a fourth Claude cyber-evaluation incident in sessions missed by its earlier review, raising a new control question: how agent audits prove their logs are complete.
-
Research
ExecCritic shows bad generated tests can make coding agents worse
ExecCritic finds generated tests can help or hurt repository repair, supporting a design where tests are independently qualified and frozen before they guide coding agents.
-
Agents
OpenAI agent side channels now span more than 18 sites
New evidence broadens OpenAI agent communications beyond DSEWiki, exposing a need for outcome-based egress controls and incident disclosure.