Expired50% confidenceFactExact time
Large model APIs universally support the "system prompt" (System Prompt) mechanism, corresponding to messages with the role field set to "system" in API requests, which carries the highest instruction priority and can be used to spoof model identity.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/4/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/2/2026(expired)
Sources
Exposing the 'Claude Fable 5' Scam: A Guide to Identifying Fake AI Products
bilibiliGPT官方车队7/2/2026
Related Claims
Unverified大模型基础调用入门的核心是API调用稳定和提示词精准72% similarUnverified大模型 API 是指 OpenAI、Anthropic、百度文心、阿里通义等厂商将训练好的大型语言模型能力以接口形式对外开放，企业只需调用接口即可获得文本生成、理解、推理等能力72% similarUnverifiedModel Provider是AI工程中适配器模式（Adapter Pattern）的典型应用，通过定义标准接口封装不同大模型服务商的API差异70% similarUnverifiedMainstream large model APIs all provide HTTPS-based REST API interfaces, allowing developers to invoke model capabilities via standard HTTP POST requests with only an API key, making the technical barrier extremely low.69% similarUnverifiedMost mainstream large models currently support tool calling, but the implementation method and API differ for each model.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/75357API
curl https://kongchang.com/api/v1/knowledge/claims/75357MCP
get_claim(id=75357)