AgentOps

A lightweight artifact for documenting agent roles, tools, memory, communication, monitoring, governance, versioning, and evaluation.

A companion concept for defining responsibilities, allowed actions, escalation rules, and operational constraints for AI agents.

A governance concept for recording agent versions, decisions, tool use, evaluations, and operational evidence over time.

The rapid adoption of LLM-based agents raises new challenges for transparency, reproducibility, and governance. Existing artifacts (Model Cards, Data Sheets) document models and data, but there is no standard for the operational characteristics of agents. We introduce Agent Cards, a lightweight artifact that captures roles, memory, tool integrations.