SourcesAgentShield First Open Benchmark 6 Agent Security Products 537 TestsAgentShield·high signalXBlueskyLinkedInCopy linkFirst open benchmark testing 6 agent security tools across 537 cases. Tool abuse detection is weakest category. Market optimizing for wrong threat.SourceSource pageAgentShield↳ Follow the threadPolicy dependency / Stack layerCherry Studio 2.0 Ships a Built-In Agent Runtime That Reuses Your Claude and Codex SubscriptionsGitHubStack layer / Threat patternagentacct Builds a Local Cost-and-Action Ledger for Coding Agents With No Login and No TelemetryGitHubStack layer / Threat patternPattern: Agent Cost Observability Is Consolidating Into Its Own Tool CategoryGitHubStack layer / Threat patternopen-kritt Open-Sources Agent-Orchestrated Vulnerability Hunting on the Premise That Whole-Repo Scanning FailsGitHubStack layer / Threat patternclaude-red Packages Offensive Security Methodology as SKILL.md Files — 457 Stars in Two DaysGitHubStack layer / Threat patternToolport Topped Out at #5 on Product Hunt With a Free MCP Gateway That Cuts Tool-Definition Overhead 96%GitHub tsouth89/toolport (corroborated by Product Hunt August 8 leaderboard, 102 upvotes, and toolport.app)Stack layer / Threat patternAgno patches an MCP flaw where a model could smuggle `tool_name` and execute any tool, bypassing every approval gateGitHub (agno-agi/agno)Stack layer / Threat patternNativ Turns a Mac Into an Anthropic- and OpenAI-Compatible Local Server With Prewired Configs for Ten Coding AgentsGitHub