Unverified50% confidenceFactExact time
Attackers tamper with a model's chat template by calling the create API, permanently injecting malicious instructions into the System prompt; this change is stored in the model's Modelfile and is invisible to the client and undetectable by standard tools.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedPrompt injection attacks can use carefully crafted user inputs to induce the model to ignore the constraints of the system prompt.75% similarVerifiedSystem prompts are a layer of instructions preset by vendors and invisible to users, injected into the model context before the user's conversation begins, used to shape the model's behavioral boundaries, response style, and safety policies.73% similarUnverifiedA System Prompt is a hidden instruction layer deployed before the user-visible conversation, pre-configured by developers to set the boundaries of model behavior, and remains in effect throughout the entire conversation.72% similarUnverified在大语言模型交互架构中，系统提示词通常由平台开发者预设，用于定义模型的身份、能力边界和行为规范，对终端用户不可见69% similarUnverified当模型输出工具调用语法后会停止响应，Harness执行工具并将输出追加到对话历史末尾，然后向同一模型重新发起请求68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/915604API
curl https://kongchang.com/api/v1/knowledge/claims/915604MCP
get_claim(id=915604)