Unverified50% confidenceEventExact time
Zenity researchers injected prompts into Salesforce Agentforce through the public Web-to-Lead form and bypassed the URL filtering rules Salesforce added after Noma Security's earlier discovery.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedBy 2024, security researchers had demonstrated indirect prompt injection attacks that hijack RAG system outputs by embedding hidden instructions in web pages58% similarUnverifiedThe method for verifying a tool's machine readability is to have an unfamiliar agent answer five questions using only the machine URL: the service's purpose, limitations, pricing, trust credentials, and invocation path.56% similarUnverifiedIn 2023, researchers demonstrated hiding malicious instructions in invisible text areas of web pages, successfully hijacking a GPT-4-based browser agent into performing unauthorized operations.55% similarUnverified2023年安全研究员Riley Goodside与Johann Rehberger等人演示了通过网页内容、邮件正文劫持GPT-4插件和AutoGPT的完整攻击链55% similarUnverifiedPublicly available information does not disclose the specific technical implementation of Agent Looker (such as whether it integrates via API, browser extension, or middleware proxy) nor details such as its trust judgment accuracy and false positive rate.54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1025911API
curl https://kongchang.com/api/v1/knowledge/claims/1025911MCP
get_claim(id=1025911)