Staying Safe with goose
Keep your goose workflows safe with built-in security features and best practices
Staying Safe with goose
Understanding security risks and following best practices can help you secure your workflows with goose.
📚 Documentation & Guides
- [Adversary Mode](/agents/goose/guides/security/adversary-mode/)：An independent agent reviewer that silently watches tool calls to protect you in case the agent goes rogue.
- [Prompt Injection Detection](/agents/goose/guides/security/prompt-injection-detection/)：Detect and prevent potentially harmful commands before they run.
- [Classification API Specification](/agents/goose/guides/security/classification-api-spec/)：API specification for self-hosting ML-based prompt injection detection endpoints.
📝 Featured Blog Posts
- [How to Determine If An MCP Server Is Safe](https://goose-docs.ai/blog/2025/03/26/mcp-security)：Before you plug your agent into just any MCP server, here's how to check if it's actually safe.
- [Securing the Model Context Protocol](https://goose-docs.ai/blog/2025/03/31/securing-mcp)：Building secure and capable AI integrations with Model Context Protocol (MCP) at Block.