repetition 5

Thinking-loop guard for dsh: observes the llm/stream waterfall and breaks a model that degrades into a loop. Detects two per-call shapes (reasoning-only calls, restated-material calls) and re-fires instead of latching; plus two mid-stream breakers that en

Thinking-loop guard for dsh: observes the llm/stream waterfall and breaks a model that degrades into a loop. Detects three per-call shapes (reasoning-only calls, restated-material calls, self-repeating low-entropy reasoning) and re-fires instead of latchi

Thinking-loop guard for dsh: observes the llm/stream waterfall and breaks a model that degrades into a loop. Detects two per-call shapes (reasoning-only calls, restated-material calls) and re-fires instead of latching; plus two mid-stream breakers that en

Dead-loop rescue for DeepSeek Harness: detects model repetition mid-stream, prunes the loop from the committed message, and resumes the task automatically.

DeepSeek Harness 重复输出抑制器：在 llm/stream 层拦截思维链与正文的退化性重复（ok/好的/马上 死循环），provider 无关，按模型匹配。