Unverified85% confidenceFactExact time
LLM在代码生成场景中引用软件包时，是基于上下文语义和训练数据模式合成一个看似合理的包名，而非从实际的包注册表中检索
1
Sources
85%
Confidence
Long-term
Relevance
8/2/2026
First Seen
Sources
Claude Package Hallucination Vulnerability: AI Generates Fake Dependency Packages That Steal Real API Keys
Related Entities
Related Claims
UnverifiedThe most useful scenario for LLMs is having them query and understand code, rather than generating new code.76% similarUnverifiedNo-code platforms pre-package LLM technical details, simplifying LLM calls into a configurable 'behavior node'.68% similarUnverified项目级规格配置文件通过在每次会话启动时自动注入项目规范，人为构造持久化上下文，弥补LLM没有跨对话持久记忆的天然局限67% similarUnverifiedLLM 工具支持模板和系统提示词功能67% similarUnverifiedSourcery combines a rule engine with LLMs and is capable of generating directly applicable code refactoring patches.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/679877API
curl https://kongchang.com/api/v1/knowledge/claims/679877MCP
get_claim(id=679877)