Agent Workflows
Writing filed under Agent Workflows.
Fresh context review is not bureaucracy
Verifier agents catch different failures when they are not soaked in the builder rationale.
AI SDK 6 made tool approval mainstream
Approval buttons need context
An approval step only helps if the human can see the plan, the data, and the rollback story.
Claude 4 made long-running agent work a planning problem
Codex made parallel agent work feel normal
Handoffs are part of the agent
MCP turned agent tools into an interface
Why I like small agent tools
Small tools are easier to permission, easier to review, and harder for an agent to misuse quietly.
Agent runbooks need failure examples
Runbooks for agents get better when they include the mistakes the agent is expected to make.