The UnboundCompute blog
Writing on web security
-
postMessage Vulnerabilities: When Cross Origin Messages Turn Into XSS
postMessage vulnerabilities let any site message your page or leak your data. Learn the origin check and targetOrigin mistakes, the XSS chain,…
-
Client Side Path Traversal: When the Browser Sends Your Fetch Somewhere Else
Client side path traversal lets attacker input redirect a browser fetch to a different endpoint. Learn how it works, how it chains,…
-
Cross Site WebSocket Hijacking: The CSRF of WebSockets
Cross site WebSocket hijacking lets a malicious page open an authenticated socket to your logged in app. Learn how it works and…
-
System Prompt Extraction: Why Keeping the Prompt Secret Is Not Security
System prompt extraction gets an LLM app to reveal its hidden prompt. Learn how it works and why secrets should never live…
-
Denial of Wallet: When Attackers Run Up Your AI Agent’s Bill
A denial of wallet attack keeps your AI agent running and drives the bill up instead of taking it offline. Learn how…
-
MCP Tool Shadowing: When One Server Hijacks Another’s Tools
MCP tool shadowing lets a malicious server collide with a trusted tool name or rewrite how it is used. Learn how it…
-
ASCII Smuggling: Invisible Unicode Prompt Injection That Humans Cannot See
ASCII smuggling hides instructions for an LLM in invisible Unicode so the model obeys while the human sees clean text. How the…
-
Slopsquatting: When Attackers Register the Packages AI Hallucinates
Slopsquatting is when attackers register the fake package names that AI invents in code. Learn how the attack works and how to…
-
The MCP Rug Pull: When an Approved Tool Changes After You Trust It
The MCP rug pull attack swaps an approved tool’s definition after you trust it. Learn how the time based mutation works, how…
-
LLM Data Exfiltration Through Markdown Image Rendering
LLM data exfiltration hides secrets in rendered markdown images and links. See how the browser leaks them on load, how to detect…