Tool loop exhaustion
A failure state where an AI agent gets stuck repeatedly executing the exact same incorrect tool actions until it runs out of its allowed token budget or context window.
What it is
Tool loop exhaustion occurs when an agent's reasoning fails in a loop. For example, it might write a test, see it fail, rewrite the exact same code, fail again, and repeat indefinitely. Because each step adds history to the context window, the agent quickly exhausts its token budget or context limit, causing the workflow to crash.
When you would use it
You design guardrails, hard limit caps, and human-in-the-loop checkpoints specifically to detect and prevent tool loop exhaustion in production orchestration systems.
Common operations
- Not applicable.
Related terms
Where this is taught
No learning path uses this term yet. Browse the Learning Atlas for guided sequences through related ideas.