CONCEPT

Harness

The scaffolding layer around a language model that turns it into a working agent — tools, prompts, memory, execution environment, and safety checks. A common formulation holds that the model is a small part of an agent's practical capability and the harness the larger part, making it a distinct engineering object.

category

Agent abstraction layer (tools, prompts, memory, security)

role

Layer that DeepSeek sums up with the equation "Agent = Model + Harness": the model is *"the soul of the agent"*, the harness is what lets it understand its environment, use tools, and keep working under real conditions. DSH offers an operational definition of it through plugin composition and a single log, rather than a feature catalog

equation

Agent = Model + Harness; model ~10%, harness ~90%

Fiches (3)