Unverified50% confidenceSolutionExact time
Outside's system prompt treats user speech as untrusted descriptive data rather than instructions, in order to resist prompt injection attacks.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSystem prompt-based solutions are essentially natural language conventions and can easily be bypassed by jailbreak attacks, prompt injection, or unintended model behavior.77% similarUnverified提示注入攻击的根源在于LLM在推理时无法从语义层面可靠区分「系统指令」与「用户数据」，两者均以自然语言形式存在于同一上下文窗口中75% similarUnverified用户可以通过提示注入和语义改写等手段绕过内容检测，使实际拦截效果是概率性的而非绝对的74% similarVerifiedPrompt injection induces agents to deviate from the original task intent by embedding text disguised as instructions within external content processed by the agent.72% similarUnverifiedThe fundamental logic of defending against indirect injection is the separation of data and instructions: retrieved content should be confined to a read-only data zone, and hierarchical isolation can be established by placing external content in the tool role rather than the user role.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1003365API
curl https://kongchang.com/api/v1/knowledge/claims/1003365MCP
get_claim(id=1003365)