Unverified50% confidenceFactExact time
A large model by itself does not solve any end-to-end problem: it does not remember context, does not call APIs, does not read databases without tools, and does not actively retry failed operations.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Harness Engineering: The LLM Is the Brain, the Harness Is the Body
hackernewshackernews10/19/2026
Related Entities
Related Claims
UnverifiedThe model itself is a pure inference component, with no network access capability, no file system, and no ability to perform any side-effecting operations.72% similarUnverified所有LLM本质上都是高级的自动补全，模型本身无法编辑文件、操作数据库、连接外部服务或上网，只能生成文本68% similarUnverifiedLarge models are stateless and do not retain the previous round of context; all history and tool definitions must be reassembled by the client before being sent.67% similarUnverifiedNative large models cannot interact with external systems and cannot query databases, call third-party interfaces, or access the file system.67% similarUnverifiedBrainstorm does not solve model capability problems, nor does it replace code review.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1068414API
curl https://kongchang.com/api/v1/knowledge/claims/1068414MCP
get_claim(id=1068414)