Unverified50% confidenceFactExact time
In the Agent event debugging page, platform tools use Mock: the Agent actually calls the tools, but platform actions such as replying and sending are simulated and no real messages are sent.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
Unverified提示注入是针对操作型Agent的独特新型攻击向量，恶意网页可通过视觉元素或隐藏文本向Agent注入指令，如何区分真实指令与伪造指令仍是尚未完全解决的开放问题72% similarVerifiedIndirect prompt injection refers to attack instructions hidden in external web pages, documents, or API responses accessed by an agent, where the malicious instructions are activated when the agent retrieves and injects them into the context.70% similarUnverifiedAgent系统中常见的触发器类型包括用户消息、定时任务（Cron表达式）、外部系统回调（Webhook）、Agent间协作和数据变更事件70% similarUnverifiedIn agent scenarios, prompt injection can translate into real actions, such as reading private data, calling send interfaces, or modifying production configurations, so mitigations must be architectural.69% similarUnverified在Agent框架中，模型幻觉出的URL会直接触发真实的网络请求，使幻觉风险从文本层面延伸到行为层面69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1066155API
curl https://kongchang.com/api/v1/knowledge/claims/1066155MCP
get_claim(id=1066155)