loop-guard

A project-agnostic multi-agent collaboration protocol for the DeepSeek Harness: loop guards that stop runaway agent loops, a six-field handoff format, risk-tiered review routing, and a fixed verify→commit→push→review→close sequence. One skill instead of the same rules copy-pasted into every agent's instructions.

Detect & break thinking-chain loops in DSH agents at the turn boundary (no proxy). Ported from ollama-loop-guard.

DeepSeek Harness runtime guard for tool-call and assistant-output loops