SafeClaw Setup FAQ

How do I install SafeClaw?

Run npx @authensor/safeclaw in your terminal. This downloads and executes the SafeClaw setup process. No global installation is required. The command launches the setup wizard, which guides you through policy creation, rule configuration, and control plane connection. The entire process is designed to take minutes, not hours. See also: What Is SafeClaw? FAQ.

What are the system requirements?

SafeClaw requires Node.js installed on your machine. It runs on macOS, Linux, and Windows. SafeClaw is written in TypeScript strict mode with zero third-party dependencies, so there are no additional packages to install beyond the SafeClaw client itself. The client runs locally and evaluates policies in sub-millisecond time with minimal memory footprint.

Do I need Node.js?

Yes. Node.js is the only prerequisite for running SafeClaw. If you already have Node.js installed (which is common for developers working with AI agents), you can run npx @authensor/safeclaw immediately. The npx command is bundled with Node.js and handles downloading the package automatically.

How long does setup take?

Initial setup typically takes under five minutes. The setup wizard walks you through each step: creating an API key, defining your first policy, and configuring rules for file_write, shell_exec, and network actions. No manual configuration file editing is required. Users who want more control can edit policy files directly after the wizard completes. See also: Policy Engine FAQ.

Do I need to write config files?

No. The setup wizard and browser dashboard handle policy configuration visually. You can create, edit, and manage rules without writing JSON or YAML by hand. For advanced users, policies are stored in standard configuration files that can be version-controlled and edited programmatically. Both approaches are fully supported.

How do I create my first policy?

The setup wizard creates your first policy during installation. It asks what actions your agent needs to perform and generates appropriate allow rules. All other actions are denied by default. You can refine the policy at any time through the browser dashboard. SafeClaw also provides simulation mode for testing policies before enforcing them. See also: Policy Engine FAQ.

What is the setup wizard?

The setup wizard is an interactive guide that launches during npx @authensor/safeclaw installation. It walks you through: selecting action types to gate (file_write, shell_exec, network), defining allow rules for your agent's legitimate operations, generating a deny-by-default policy, and connecting to the Authensor control plane. The wizard is designed for users with no prior security tooling experience.

Can non-technical users set it up?

Yes. SafeClaw was built with non-technical users in mind. The setup wizard uses plain language, not security jargon. The browser dashboard provides a visual interface for policy management — no terminal commands required after initial installation. This is a key differentiator from tools like Docker or OS-level sandboxing, which require significant technical expertise. See also: SafeClaw vs Alternatives FAQ.

How do I get an API key?

API keys are generated during the setup wizard or through the SafeClaw browser dashboard at safeclaw.onrender.com. Keys are issued on the free tier with a 7-day renewable period. No credit card is required. Keys authenticate your SafeClaw client with the Authensor control plane, which receives only action metadata — never your API keys or sensitive data. See also: Privacy and Trust FAQ.

What is the free tier?

SafeClaw's free tier provides full access to all features: action-level gating, the policy engine, the tamper-proof audit trail, the browser dashboard, the setup wizard, and simulation mode. Free tier API keys are valid for 7 days and are renewable at no cost. No credit card is required to sign up. Visit safeclaw.onrender.com or authensor.com to get started.