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 Safeguard CLI is a single static binary that can scan and write code. How to run it inside a cloud or local agent sandbox, and what network it needs.
When a Go repo has a vendor directory, builds read it instead of the module cache. Here is what that means for SBOM generation and which file to check.
Track blocked actions, approval rates, agent-added dependencies, review depth, and credential age. Six metrics that show whether agent controls work.
A 45 minute agenda for a security-led session on an AI coding tool: data path, execution, reach, logs, administration, and the live tests to request.
Cloud sessions run the agent in an isolated environment with controlled credentials and network. Local sessions stay on your machine. Compare the two.
Yes. Route your end-to-end tests through a scanning proxy and get passive findings for free, then add targeted active scans. Steps, benefits and limits.
Start agents in read-only or plan mode, allow edits in a sandbox, and keep command and network actions gated. A decision guide for choosing a mode by task.
Everything you send plus the reply counts toward the context window: prompt, history, tool results, tool definitions and reasoning. See why bills grow.
Yes. Files that steer a coding agent, grant it permissions, or run hooks change what it does. Put them under ownership, review, and change history.
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.
