Unverified50% confidenceFactExact time
DSH提出核心公式：智能 = Model + Harness
1
Sources
50%
Confidence
Long-term
Relevance
8/21/2026
First Seen
Sources
Related Claims
UnverifiedA DSH project typically contains system prompts, Skills, and the definition of the Agent.71% similarUnverifiedIn the energy storage agent architecture, the LLM large model is responsible for thinking, analysis, and decision-making; MCP is responsible for connecting to industrial systems; and Skills is responsible for executing specific tasks.70% similarVerifiedThe core architecture of an agent can be summarized as: Agent = Large Language Model (LLM) + Tools + Memory + Task Scheduling.69% similarUnverifiedSKILL.md 中显式描述工作流可以将模型行为锁定在可预期的执行路径上，提升任务完成质量69% similarVerifiedThe core formula of Harness Engineering is: Harness (specification) + LLM (large language model) = Agent.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/784000API
curl https://kongchang.com/api/v1/knowledge/claims/784000MCP
get_claim(id=784000)