AgentsEntro MCP Audit Plugin Claude CodeEntro Security·high signalXBlueskyLinkedInCopy linkEntro Security ships MCP Audit Plugin for Claude Code. Real-time visibility into agent sessions, tool invocations, MCP requests. In-house SLM classifies session intent.SourceSource pageEntro Security↳ Follow the threadPolicy dependency / Stack layer'Do this as quickly as possible' repeatedly got a Claude session flagged by a corporate security directorr/ClaudeAIPolicy dependency / Stack layerClaude Code adds omitClaudeMd so subagents can run without inherited CLAUDE.md, and a sha256 plugin-install accept flagGitHubPolicy dependency / Stack layerA Fine-Tuned RoBERTa-Large Permission Gate Matches Claude Haiku 4.5 at Deciding What an Agent May ToucharXiv 2609.15422Policy dependency / Stack layermcp-airlock is a stateless governance proxy that sits in front of MCP servers with policy, dry-run and human confirmationGitHubPolicy dependency / Stack layerOrdewell uses a read-only coding agent as planner, then spawns per-task agent sessions gated by a marker-based verdict engineGitHubPolicy dependency / Stack layerA Bun packaging quirk left a vulnerable undici in Cline after the CVE was supposedly remediatedGitHubPolicy dependency / Stack layerConverting GUI Trajectories Into Replayable MCP-Style Calls Instead of Unstructured MemoriesarXiv 2609.16635Policy dependency / Stack layerVercel's Is Agentic audit now scores sites by type, with an opt-in meta tag for docs, business, app or commerceVercel Changelog