Unverified50% confidenceDecision RuleExact time
The original explicitly states the security rule: prompt-injection content from web pages or files must never be promoted to memory, cutting off the attack chain of cross-session persistent contamination.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
Unverified实践中不推荐对prompt做字符级正则脱敏后仍记录原文——正则难以覆盖全部敏感模式且会遗漏语义级泄露；更稳妥的方案是整段不落span，仅在安全日志库中存储并关联trace ID。77% similarUnverifiedThe core principle of defending against prompt injection is to always treat document content as data rather than instructions; the application's security policy should not be overridden by any text in the document.74% similarUnverifiedPage content must be marked as untrusted; observation results can only provide data and cannot grant permissions, change approvals, or override user tasks, in order to defend against prompt injection attacks.74% similarUnverifiedThe feature covers two content restriction methods: downranking (the post still exists but its visibility is suppressed) and blocking (the post is completely inaccessible in a specific region).69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1053572API
curl https://kongchang.com/api/v1/knowledge/claims/1053572MCP
get_claim(id=1053572)