Verified80% confidenceFactExact time
Indirect prompt injection attacks plant hidden instructions within normal business data to induce Agents into executing unauthorized operations, representing a security risk at the semantic level rather than the code level.
6
Sources
80%
Confidence
Long-term
Relevance
7/6/2026
First Seen
Sources
Related Claims
Unverified将敏感信息置于模型可访问范围之外是防御 prompt 注入类攻击的有效隔离措施80% similarUnverified涉及代码执行的任务中某些异常信息（如动态代码求值报错）可能被安全系统误读为威胁信号79% similarVerifiedPrompt injection attacks induce AI agents to execute malicious operations by embedding text disguised as system instructions in code comments, documentation, web content, or even file names.78% similarUnverified在具备工具调用能力的 Agentic 系统中，一次成功的提示注入可能导致数据泄露、恶意代码执行甚至供应链攻击78% similarUnverifiedLLMs are vulnerable to prompt injection attacks, where malicious web pages or files can embed hidden instructions to induce AI to perform unintended actions.78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/113192API
curl https://kongchang.com/api/v1/knowledge/claims/113192MCP
get_claim(id=113192)