Unverified50% confidenceOpinionExact time
Browy's security model assumes it will always encounter malicious pages, because it reads pages written by strangers and is logged in as the user.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAttackers can hide malicious instructions in web pages, documents, or even images, and when an agent reads this content it may mistake the instructions for legitimate commands and execute them69% similarUnverifiedMalicious websites can embed hidden instructions in their pages to induce AI to perform unintended purchasing actions.68% similarUnverifiedOne-click exposure of a model from a Notebook as a public HTTPS endpoint requires handling security issues such as authentication, rate limiting, and data privacy.68% similarUnverifiedThe HTML preview feature of the Claude Code Desktop application has a vulnerability that attackers can exploit using scripts hidden in preview files to read local data such as SSH private keys, AWS credentials and GitHub tokens.66% similarUnverifiedAgents lacking isolation face indirect prompt injection attacks when scraping local HTML documents or websites containing maliciously hidden prompts, as language models cannot reliably distinguish legitimate commands from hidden payloads embedded in documents.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1071580API
curl https://kongchang.com/api/v1/knowledge/claims/1071580MCP
get_claim(id=1071580)