Agent harnesses are becoming the dominant way to build agents, and they are not going anywhere. These harnesses are intimately tied to agent memory. If you used a closed harness - especially if it’s behind a proprietary API - you are choosing to yield control of your agent’s memory to a third party. Memory is incredibly important to creating good and sticky agentic experiences. This creates incred...
The strongest version of this narrative is that agent harnesses are evolving into the backbone of AI systems, with memory as their most strategic component. The argument that closed harnesses create lock-in is well-supported by examples like Anthropic’s Claude Managed Agents and OpenAI’s proprietary compaction. The call for open-source alternatives like Deep Agents is compelling, as it aligns with broader movements toward user sovereignty in technology. However, the piece assumes that open-sourc...