AI Security
What an MCP server must do on every tool call
The MCP tools spec says servers must validate inputs, enforce access control, rate limit and sanitize outputs. A practical checklist for building a tool.
Oct 18, 20264 min read
Resources
Supply Chain Security, in plain English.
Deep dives, practical guides, and incident analyses from engineers who build Safeguard. No fluff, no vendor FUD — just what you need to ship secure software.
Filter
Articles
RSS feedAI Security
MCP HTTP servers: Origin checks and DNS rebinding explained
Local MCP servers over HTTP can be reached from a website through DNS rebinding. The spec requires Origin validation and recommends binding to localhost.
Oct 18, 20263 min read
AI Security
MCP confused deputy: how proxy servers skip user consent
An MCP proxy with a static client ID and a consent cookie can hand an attacker an authorization code. Learn the four conditions and the required fixes.
Oct 18, 20263 min read
AI Security
MCP client SSRF: defending OAuth discovery from hostile servers
A malicious MCP server can point OAuth discovery URLs at internal services and cloud metadata. The spec lists the defenses clients should apply.
Oct 18, 20263 min read
AI Security
MCP clients: validate authorization URLs to stop XSS and RCE
A hostile MCP server can send a javascript: or shell-laced authorization URL. Clients must allow only http and https and never open URLs through a shell.
Oct 18, 20263 min read
AI Security
Markdown image exfiltration: how chat output can leak data
A model told to add an image can send chat contents to an attacker's server when the page renders it. How the leak works and how to close it.
Oct 18, 20263 min read
AI Security
Mapping a RAG app to MITRE ATLAS techniques, step by step
How to map a retrieval-augmented generation app to the ATLAS techniques that apply: recon, poisoning, false entries, credential harvesting and data theft.
Oct 18, 20264 min read
AI Security
LLM security review: five myths and what is actually true
Five common beliefs about LLM security review, such as a better prompt fixing everything, and what OWASP, the NCSC and Anthropic guidance say instead.
Oct 18, 20263 min read
AI Security
Letting an LLM review pull requests from strangers: the risks
A pull request from a stranger can contain text aimed at your AI reviewer. How indirect prompt injection reaches review bots and how to limit the damage.
Oct 18, 20263 min read
Stay informed
Weekly insights on software supply chain security, delivered to your inbox.
Self-healing security runs on Safeguard.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.