Agent Stop Conditions

Use when designing an agent, reviewing its orchestration logic, or investigating a runaway loop. Unbounded loops, missing human gates, or silent retries are failures. Encode when to stop and escalate before the agent runs, not after it burns budget or corrupts state.

frankxai Updated 47 repo stars

File contents

frankxai/skills commit e98494fc80

Frequently asked questions

npx get-skill add frankxai/agent-stop-conditions