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.
OpenAI shipped ChatGPT Atlas in October 2025 and admitted by December that prompt injection in AI browsers may never be fully solved. Defenders need a posture, not a patch.
LayerX's October 2025 CometJacking attack siphoned Gmail and Calendar via one Perplexity Comet click. The browser-as-agent design is the new blast radius.
A Cursor user's Supabase MCP server was tricked by a support ticket into exfiltrating an integration_tokens table. The bug was not in MCP. It was in the trifecta.
Aim Security's CVE-2025-32711 exfiltrated Microsoft 365 Copilot data via a single crafted email. The XPIA classifier failed, CSP let attackers through, and CVSS 9.3 followed.
A single crafted email could exfiltrate data from Microsoft 365 Copilot without a user click. We walk the attack chain, the patch, and the lessons for agent operators.
Invariant Labs showed that a malicious GitHub Issue could hijack any MCP-connected agent into leaking private-repo contents. The architecture, not a bug, is the problem.
Trail of Bits coined 'line jumping' for prompt injection delivered through MCP tool descriptions on connection. It bypasses every tool-invocation guardrail by design.
The OWASP Top 10 for LLM Applications 2025 added System Prompt Leakage and Vector/Embedding Weaknesses, and elevated Sensitive Information Disclosure to #2. Here is the defender view.
OWASP published its first Top 10 for LLM Applications on August 1, 2023. Here is what it covers, where it overreaches, and how to use it on real systems.
Weekly insights on software supply chain security, delivered to your inbox.