AgentsAgentShield Benchmark: First Open Test of 6 Agent Security VendorsGitHub·high signalXBlueskyLinkedInCopy link537 test cases, 8 categories. Scores range 39-98. Critical: providers catching 95%+ prompt injections miss most unauthorized tool calls. Provenance verification nearly absent.SourceSource pageGitHub↳ Follow the threadStack layer / Threat patternBeacon logs agent sessions across Claude Code, Cursor, Codex and 20+ other harnesses and turns corrections into reviewed memory served over MCPGitHub TrendingStack layer / Threat patternMiMo Code 0.1.15 adds a tool-call sequencing gate: only read/search calls run in parallel, and side-effecting calls run in orderGitHub TrendingStack layer / Threat patternOllama 0.34.4-rc0 applies structured outputs in one pass on thinking modelsGitHubStack layer / Threat patternweave-os/router gained 192 stars today: a sub-50ms embedding router that sits in front of Claude Code, Codex, opencode and CursorGitHub TrendingStack layer / Threat patternBuilder.io's Agent-Native framework picks up 609 stars in a dayGitHubPolicy dependency / Threat patternCodex now enforces network policy on every redirect, streamed body read and open WebSocketGitHubStack layer / Threat patternCline 4.1.20 runs sibling sub-agents' tool calls in parallel and restores hook-injected contextGitHubStack layer / Threat patternQwen Code adds a hidden attestation-only worker process for its managed runtimeGitHub