Unverified50% confidenceFactExact time
Recovery paths can stack in a chain: after a reactive compact succeeds it may hit max_tokens, the transition becomes max_output_tokens_escalate, and if reached again it becomes max_output_tokens_recovery.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe main loop translates stop_reason into the next action decision: tool_use enters the tool execution branch, max_tokens triggers max_output_tokens_escalate or max_output_tokens_recovery, and prompt_too_long maps to reactive_compact_retry.72% similarVerifiedReAct is more flexible and can dynamically adjust strategies, while Plan-and-Execute is more stable in long task chains, reducing global deviation caused by local errors.65% similarUnverifiedRetry occurs within a single execution for a temporary failure of an individual node, while re-run is the re-execution of the entire workflow after the execution has terminated.64% similarUnverifiedstate.transition.reason is a union type with seven values: next_turn, collapse_drain_retry, reactive_compact_retry, max_output_tokens_escalate, max_output_tokens_recovery, stop_hook_blocking, token_budget_continuation.63% similarUnverifiedBuilding a fallback chain that automatically switches to the next provider and retries when a provider returns 429 can improve the overall availability of an application63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1028900API
curl https://kongchang.com/api/v1/knowledge/claims/1028900MCP
get_claim(id=1028900)