RedditAI Agents as Identity Dark Matter: 70% Enterprises Running Agents UngovernedWweb·medium signalTeam8 CISO survey: 70% enterprises run AI agents in production, 23% planning 2026 deployments. Agents invisible to IAM. Five dark matter risks: over-permissioned access, untracked usage, static credentials, regulatory blind spots, privilege drift. MCP adoption accelerates the gap.↳ Follow the threadPolicy dependency / Stack layerPattern: this week's enterprise agent news is dominated by runtime governance and cost controlAI Agent StoreStack layer / Threat patternGhostcommit hides prompt injection in PNG files to make AI coding agents exfiltrate repo secretsSecNewsPolicy dependency / Stack layerPattern: A Provider-Abstraction / BYOK Layer Is Hardening Beneath Agent HarnessesGitHubStack layer / Threat patternFriendly Fire: auto-approve coding agents run attacker code during security scansSecurityWeekStack layer / Threat patternFeatured launches an MCP server letting PR-team agents run inside scoped user accountsAI Agent StorePolicy dependency / Stack layerFirst Recon AI ships AI Security Runtime GA — inline policy enforcement on every agent-to-agent and agent-to-tool callBusiness WireStack layer / Threat patternScan for MCP prompt injection at a centralized control plane, and treat every tool response as untrustedObot AIPolicy dependency / Stack layerCodenotary ships AgentMon 3 for agent runtime policy and compliance loggingAI Agent Store