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.
On a Claude Team or Enterprise plan, an Owner adds the Safeguard connector once and members connect it themselves. Here is the order of steps and why.
Several ways to add an MCP server in VS Code, which config file each uses, how workspace trust works, and how to sandbox and debug a server safely.
The exact claude mcp add commands for remote HTTP, local stdio and JSON config, plus how to check a new server is connected and how to remove it.
Where Cursor keeps MCP config, how stdio and remote servers differ, how OAuth redirects work, and what the tool approval prompt shows you before it runs.
How to add a remote MCP server as a custom connector in Claude, who can do it on each plan, why it must be public, and which cautions to follow.
The scan command takes one target: a directory, a container image, a Git repository or an SBOM. Here is when each one fits and what each one misses.
Match your editor to the right Safeguard extension: VS Code, JetBrains IDEs, or Open VSX for Cursor, Windsurf and Codium. Versions as of 19 Oct 2026.
A duration clause on an alert delays all incidents equally and resets on brief recovery. See why it fails and what to use instead.
A new Safeguard workspace exposes six read-only MCP tools by default. Every other tool stays off until an administrator enables it. Here is what to do.
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.
