
custodyn. — Agent Trust Platform
Add policy enforcement, human approvals, and audit logs to any AI agent via SDK integration.
@4tul_tiwari · X
I built an AI agent security platform — policy enforcement, human approval queues, tamper-evident audit logs — embedded directly inside agent code via SDK. LangChain, CrewAI, AutoGen, MCP, Claude Code, OpenAI Agents SDK — all supported. Live at


