Unverified50% confidenceFactExact time
Typical representatives of the "LLM as high-level planner" architecture are works such as SayCan and Code as Policies.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThis architecture positions the LLM as the task planning layer rather than directly generating low-level control signals, with embedded controllers (such as the ESP32 series) executing function calls sequentially.70% similarUnverifiedStructured output is a key capability of LLMs in enterprise process integration, achievable by constraining output format through system prompts or by calling OpenAI's Function Calling / Structured Outputs interface.69% similarUnverifiedCoding agents possess a closed-loop perception–planning–execution capability, making them suitable for embedding in scheduled tasks, data pipelines, and other automated workflows, forming an architectural pattern where LLMs serve as workflow nodes.67% similarUnverifiedLLM 工具支持模板和系统提示词功能66% similarUnverified后台编码智能体的核心技术依赖 LLM 的代码理解能力、工具调用（Function Calling）机制以及沙箱执行环境64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114086API
curl https://kongchang.com/api/v1/knowledge/claims/1114086MCP
get_claim(id=1114086)