Unverified50% confidenceFactExact time
Data exfiltration can be achieved via image tags: if a reply generated by the Agent contains an image tag, when the message is rendered or the client loads the image, the data is sent to the attacker with the URL request.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDecisions API input can be text, images, or a mix of both, but images must be provided as base64 inline data URLs (up to 128 per request), and hosted image links and uploaded file IDs are rejected.70% similarUnverifiedRendering-based data exfiltration refers to an attacker inducing the victim's client-side rendering engine to actively initiate network requests carrying sensitive information when parsing malicious content, with common vectors including <img src>, CSS url(), <link> preloading, and Markdown image syntax, and the core of defense being Content Security Policy (CSP).70% similarUnverifiedPublishing a post with an image via the LinkedIn API usually takes two steps: first upload the image through the registered upload endpoint and obtain the image asset URN, then call the create post endpoint to submit the author identity, text, and image asset URN combined.69% similarUnverified该分镜工具支持在分镜格中输入@符号快速引用已上传的图片68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1044844API
curl https://kongchang.com/api/v1/knowledge/claims/1044844MCP
get_claim(id=1044844)