Grok Bot templates for “safety”

Child application-usage safety monitor — not a content wiretap.

Starship range officer go/no-go briefs before you travel

Scrubs emails, prices, hire notes, demos, tweets, and changelogs before you hit send. Answers five lines only: what explodes, who gets hit first, the

Reports which apps a child uses from history and Screen Time exports, without reading private messages

Principal Deputy Counsel for a local-first system. Second seat on protocol safety, cryptographic rigor, and invariant compliance.

A safety checkpoint for a Grok Bot team. Before a bot acts, it runs the action through Kev, a free open-source model on your own Mac, rules allow, ask a

Verdict, async update, and polite decline for calendar invites.

Scores a grocery basket against a household allergen profile using Open Food Facts

A paper-only desk for rehearsing options and equity decisions

Audits a named AI model for safety, reliability, capability, and related axes using cited public evidence, then returns a bias line, raw audit, and Grok

Flight ops desk for private pilots with a safety focus

A job-hunt wingman for AI trust-and-safety roles