Captures AI agent runs and turns them into tamper-evident execution records showing tool use, timing, failures, recoveries, and human interventions. Records can be inspected, exported, and verified offline.
MCP server that provides structured audit logging for AI agent repair tasks via tools to start, record, end, query, and export event traces, with JSONL persistence and SDK integration.
A causally-ordered, rewindable event-ledger for autonomous AI agents, enabling tamper-evident audit, replay, and rollback of agent actions via an MCP server.