AI and Agents
- AI Agent Management: Role, Policy, Delegation, and Version Control Synthesis
Over six days, we explored what is needed to define, manage, and audit AI agents within an organization. Today, we bring these four tools together and show why Slack and ChatGPT fall short, and why a self-hosted Git-based system is necessary.
- Who Decides When Humans and Agents Work in the Same Team?
How should work division, approval responsibility, and organizational versioning work when a software developer and a code review agent work together? Rules live in config-as-code and every decision is logged in Git.
- What Happens When Agents Hand Off Work? A2A Delegation, Approval Flows, and Accountability
Responsibility gaps, approval veto points, and result verification mechanisms when multiple AI agents delegate work to each other. How to build a safe A2A protocol.
- Agent Identity Documents: policy.md and Operational States
How policy.md defines what an agent can do, which tools it can access, and when it can work. Understanding Draft, Active, and Passive states.
- What Is an Agent: The Tool-to-Agent Distinction and Organizational Boundaries
What is the difference between a team using a tool and a team using an agent? A tool just does things. An agent carries responsibility. Organizational boundaries, ownership, and accountability.
- Unwritten Rules: The Institutionalized Shadow AI
Does your organization have AI routines that everyone knows but nobody has written down? Undocumented responsibility is the biggest source of risk.
- AI Agent Management: Role, Policy, Delegation, and Version Control Synthesis
Over six days, we explored what's needed to define, manage, and audit AI agents within an organization. Today, we bring these four tools together and show why Slack and ChatGPT fall short when managing self-hosted, Git-based systems.
- How Do Mixed Human + Agent Teams Work? Versioning Responsibility with Git and Config-as-Code
A real team doesn't consist only of humans or only of agents. When a marketing manager, developers, and AI agents work together, who is responsible? Whose approval is needed? Where are the rules stored? Auditable team management with Git, policy files, and config-as-code.
- Who Is Responsible If Agents Are Missing From Your Organization Chart?
When AI agents aren't included in your org chart, responsibility disappears. Learn why human and agent nodes must both be mapped in your structure, and how responsible_id creates accountability.
- Agent Identity Documents: policy.md and Operational States
How policy.md defines what an agent can do, which tools it can access, and when it can work. Understanding Draft, Active, and Passive states.
- How Do Agents Hand Off Work to Each Other? Agent-to-Agent Delegation, Approval Flows, and Result Verification
One AI agent delegating work to another: approval mechanisms, responsibility chains, and operational verification. How to build accountable agent networks.
- Tool or Agent? Which Is a Defined Organizational Entity
Yesterday we saw how shadow AI becomes institutionalized without written rules. Today: What is the line between tool and agent? Why does it matter?