Boost OS
Slash commands for AI coding agents
Twenty-two commands, each one a senior-dev habit your agent runs when you type the slash. Grouped by job: build, check, memory, customize. They work in Cursor, Claude Code, Codex, and every major agent.
Build
Reach for these while making something.
- /costKnow what a feature really costs before you build itBreak down a feature or vendor across money, build time, learning curve, lock-in, and ongoing maintenance.
- /designTurn a rough idea into a plan your AI can actually buildShape a vague feature idea into a plan you can hand to /preflight before any code gets written.
- /explainUnderstand what your AI just did, in plain languageTranslate the previous output into plain language for a non-coder.
- /nextNever sit there wondering what to prompt nextDraft 2-3 builder-friendly prompts you could send next, based on the recent chat.
- /spawnSplit a big task into parallel AI agentsFan a task into parallel subagents when the work has independent parts.
Check
Spot problems before they bite.
- /auditFind and delete the dead code your AI left behindFind legacy and dead code in a bounded scope, then delete with per-file confirmation.
- /healthSee what's actually breaking in productionCheck production error logs for a service in your stack and surface a grouped report.
- /mobileFind out why your page looks broken on phonesAudit one screen for mobile readiness and hand back a numbered plan you can give to /preflight.
- /preflightCatch a bad plan before your AI writes the codeAudit a plan against the existing codebase to catch drift, bloat, redundancy, and misalignment before you start coding.
- /securityCatch leaked keys and risky routes before you deployPre-deploy safety sweep across secrets, env hygiene, dependency advisories, and risky API routes.
- /syncStop your AI from writing code for library versions you don't haveSync the agent's knowledge of installed library versions so it stops writing against outdated APIs.
- /vibeCheck your AI's work before you ship itSanity-check your uncommitted work and tell you whether you're on track or about to ship a problem.
Memory
Stop re-explaining your project every session.
- /forgetClear a saved project map when it goes staleDelete a saved subsystem reference.
- /recallLoad what your agent already learned about a part of your appRead a saved subsystem reference and pull its contents into the current chat.
- /rememberStop re-explaining your project in every new chatSave useful context about a part of the project into a reference file, from this chat or by inspecting the codebase first.
Customize
Extend and maintain the OS itself.
- /newBuild your own slash command for your projectBuild your own custom slash command for this project through a short conversation.
- /pauseTurn Boost OS off for a while without losing anythingTurn off the always-on parts of Boost OS without removing anything. /resume brings them back.
- /resumeTurn Boost OS back on after a pauseTurn the always-on parts of Boost OS back on after a pause, with guards rewired and the kernel restored.
- /uninstallRemove Boost OS from a project, cleanlyCleanly remove Boost OS from the project, including managed skills, project memory, and references. Your own commands stay.
- /updateGet the newest commands without leaving your AI chatCheck for and install Boost OS updates from any AI chat: newer commands, kernel improvements, and fixes.
- /whats-newSee what shipped since your installed versionSee what shipped since your installed version, without installing anything.
The menu
The front door to everything above.
Every command, one license
Boost OS installs all twenty commands, the always-on guards, and project memory with one line. Lifetime updates included.
From $99 · 7-day refund · license key emailed instantly