AI Tool Confused-Deputy: A Deep Dive
The confused deputy problem takes on new and subtle forms when AI agents invoke tools on behalf of users. A technical deep dive with concrete mitigations.
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.
The confused deputy problem takes on new and subtle forms when AI agents invoke tools on behalf of users. A technical deep dive with concrete mitigations.
From the CVE program funding crisis to the rise of AI-targeted supply chain attacks, 2025 reshaped the software security landscape. A comprehensive look at the year's defining events and trends.
A TOCTOU race in Docker's docker cp symlink resolution let malicious containers write to host files as root. Impact, CVSS, and fixes inside.
A poisoned model looks like any other checkpoint file. Here's how model poisoning attacks work, real incidents on Hugging Face, and how detection and provenance checks catch them.
CVE-2021-21285 lets a crafted container image crash Docker Engine before 20.10.3. Affected versions, severity, timeline, and remediation steps.
An AI bill of materials (AIBOM) inventories the models, data, and dependencies behind an AI system. Here's what it is and why enterprises need one.
CVE-2021-21284 let remapped-root containers escalate to real host root, defeating Docker's userns-remap isolation. Here's the full breakdown and fix.
Real incidents show malicious Hugging Face models evading scanners with pickle exploits and reverse shells. Here's what teams need to know before the next pull.
Prototype pollution has hit lodash, jQuery, minimist, hoek, and immer since 2018. Here's how the bug works and why it keeps coming back in npm.
Weekly insights on software supply chain security, delivered to your inbox.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.