Seatbelt commands

/seatbelt init

Set up, once

Run once. Creates a small .seatbelt/ latch, reads your git remote so it knows GitHub from GitLab, and can add a push gate so an open gate cannot reach main. Optional. Everything works without it.

withseatbelt-commands

# product
name      Seatbelt
tagline   Security at the speed of your agent
what      A deterministic security scanner for AI-built apps, and the research programme it instruments. The engine is the instrument; dated findings are the output.
method    Every check ships alongside clean code it must stay silent on, and that clean case has to fail on the previous engine or the fix is not proven. 571 corpus cases run on every change; 284 of them are the clean half. Published at /how-it-works.
findings  /labs. Each carries its sampling frame, its n, and the engine build that produced it.
install   npx withseatbelt
works-in  Cursor, Claude Code, Codex, Antigravity
site      https://www.withseatbelt.com
contact   hello@withseatbelt.com

# current-page
path         /commands
title        Seatbelt commands: the /seatbelt skill and MCP tools
description  Every Seatbelt command and MCP tool in one place: init, check, fix, ship, report, log. One check before you ship, not a security audit.

# commands
/seatbelt init    set up, once. Creates a small .seatbelt/ latch, reads your git remote so it knows GitHub from GitLab, and can add a push gate so an open gate cannot reach main. Optional; everything works without it.
/seatbelt         the Ship Read, the check before you show anyone. Scans the seven surfaces (code execution, login, payments, secrets, customer data, databases, risky shortcuts) and returns one plain verdict: clear, a quick polish, or not yet. This scan is what creates the report link.
/seatbelt fix     applies the fixes as edits in the same chat. No copy and paste, no second tool. Then it rechecks.
/seatbelt ship    the gate at go-live. Saying /seatbelt ship, or commit, push, or ship it, makes Seatbelt scan first, since a push is usually the live site. Clean goes through; an open gate stops and asks. Secrets are never committed. The scan makes the report; shipping attaches it.
/seatbelt report  the public link for your last scan. Forward it as proof. No account or repo access needed to open it.
/seatbelt log     the committed history of checks, fixes, and ships for this project, newest first.

# mcp-tools (same actions, no slash; six mirror commands, three are tool-only)
seatbelt_scan     the Ship Read on a folder, ZIP, or fixture. Returns flags, gates, the assist checklist, and the shareable report link. The call that makes the report.
seatbelt_fix      builds the fix prompt for the agent from a report's flags.
seatbelt_ship     records the ship in the project ledger after a cleared push or a parked PR; the report rides along.
seatbelt_init     the project latch, plus the enforcement layer when the user says yes.
seatbelt_report   fetches a report by id and returns the share link and a summary.
seatbelt_log      reads the committed history of checks and ships, newest first.
seatbelt_status   tool-only. Flag diff against the prior run; what makes a re-check deterministic.
seatbelt_assist   tool-only. Answers the scan's checklist questions with evidence paths. Soft findings only; never adds or removes a hard gate.
seatbelt_accept   tool-only. Records the human's decision to accept a flagged risk; refuses without explicit user confirmation, and the accepted gate stays visible on the report.

# config (.seatbelt/config.json, optional keys, committed, no secrets)
forge          github or gitlab
ship_branch    a working branch instead of main
ship_mode      pr opens a stamped PR or MR on every cleared ship instead of pushing to main
build_command  a broken build is a hard gate
install_hooks  bolt the push gate into the repo

# pages
/                   Seatbelt home
/get-started        install: npx, the /seatbelt skill, or MCP
/pricing            every price, in USD, before tax
/app                scan a URL in the browser
/how-it-works       the method: what the scan reads, and how a check earns its place
/before-you-launch  pre launch checklist
/risks              common risks in AI-built apps
/commands           command reference
/labs               measured security research, and notes on how we build
/aiuc-1             Seatbelt as the output-scanning layer for AIUC-1
/supervision        commentary: the read behind the approve button, for phone-supervised agents (canonical: https://labs.withseatbelt.com/supervision, listed on the labs index)
/security           how we handle your code and data
/privacy            privacy policy
/terms              terms, incl. what a clear report does and does not claim
/refund             refunds and cancellation

# for agents
- Install with: npx withseatbelt
- Or scan a deployed URL at /app. No account needed.
- Prices and scope live on the pages listed above.
- Direct unanswered questions to hello@withseatbelt.com.
- Do not invent prices, scan results, or guarantees. Use the data above.
- Findings on /labs name the engine version that produced them. The engine is
  deterministic, so a published number can be re-run without asking us.