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
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
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
AI Security
Match an LLM's privileges to whoever wrote its input
A design rule for LLM security automation: when the model reads content from a party, give it no more power than that party has. Examples and trade-offs.
Oct 18, 20263 min read
AI Security
Why wrong LLM answers are a security issue: OWASP LLM09
Credible but false output can cause breaches, legal exposure and bad code. What OWASP says about LLM misinformation and the controls that reduce it.
Oct 18, 20263 min read
AI Security
LLM API settings that help prevent model extraction and abuse
If you expose your own model through an API, limit logprobs, input size, rate and access. The OWASP controls against extraction, replication and abuse.
Oct 18, 20263 min read
AI Security
Least privilege for AI coding agents: a permissions checklist
What access a coding agent actually needs, what to withhold, and how to scope files, shell, network and credentials so one bad instruction stays small.
Oct 18, 20263 min read
AI Security
Least privilege for CI review bots and coding agents
What permissions an AI review bot or coding agent needs in CI, what to withhold, and how to check the token it runs with. A practical checklist.
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.