Product
Safeguard Local Runner: Agentic Security on Your Laptop
The Local Runner is a command-line agent that runs Safeguard workflows against your working tree. Think claude-code-for-security, but for supply chain.
Mar 13, 20267 min read
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 Local Runner is a command-line agent that runs Safeguard workflows against your working tree. Think claude-code-for-security, but for supply chain.
The Safeguard desktop application is 1.0 on macOS, Windows, and Linux. It brings the full workflow engine, Local Runner, and offline posture reviews to developers.
The Safeguard MCP Server is publicly available and works with Claude Desktop, claude.ai, Claude Code, ChatGPT, Cursor, Gemini, and Grok. Here is the tool surface.
Self-healing containers detect, remediate, and rebuild images when CVEs appear in their dependency closure. Here is how the GA feature works in practice.
The Gold Registry crossed 6,000 curated zero-CVE packages and images across ten ecosystems in February 2026 — and is now at 10M+. Here is what was in the original milestone, how it is built, and how to use it.
Lion 2.0 is Safeguard's compliance model. The 2.0 release adds multi-jurisdiction mapping, control-level evidence, and a new export for audit packages.
Eagle 3.0 is the classification model behind Safeguard's package, image, and secret detection. Here is what changed, what moved, and what it means for alerts.
Griffin 3.0 is now generally available. Here is what changed in the reasoning and remediation model, how it behaves in practice, and the defaults you should know.
Safeguard 5.0 introduces Griffin AI, expanded SBOM analysis, and a redesigned policy engine. Here is what is new and why it matters for your security program.
Weekly insights on software supply chain security, delivered to your inbox.