Overview
What BWorlds does for a Build, and what the bworlds CLI lets you and your coding agent do with it.
BWorlds takes an AI-built app from "it works" to a product clients can rely on. For each Build (one app a Builder ships), it audits the live app against production essentials, watches it in production, and keeps a prioritized list of what needs attention. Builders see all of this in the app at app.bworlds.co.
bworlds brings the same Build to your terminal, your coding agent, and your CI. Source code only says what an app is meant to do. The CLI adds what has actually been observed around that code, so a change is planned against evidence instead of a guess.
Who uses it
- A Builder on their own Builds. Sign in with the identity you use in the app. Every Build in your Workspace is available.
- A developer or consultant helping a Builder. The Builder invites you to their Workspace from the app. You use your own BWorlds identity and see their Builds until they remove you.
- A coding agent or an automated runner. Claude Code, Codex, or a CI job runs the same commands with a personal token that represents you. Every result is available as stable JSON.
Access is the Workspace Membership itself. There are no per-Build permissions to configure. See Access and context.
What you can do
Each row maps one BWorlds capability to the screen a Builder knows in the app and to the commands that reach it.
| Capability | In the app | From the CLI |
|---|---|---|
| Read any Audit: what passed, what needs attention, how to fix it | Audits | audit list, audit show, audit runs, audit run, audit answer, audit reevaluate, audit override, control run, control status |
| Work the Build's prioritized to-do list | Findings | findings list, findings show, findings prompt, findings verify, findings create, findings update-status |
| See what breaks in production and what your users live through | Errors, Sessions, Uptime | telemetry errors, telemetry sessions, telemetry uptime |
| Keep the Build's shared memory: how it works and what never to touch | CLI only | dossier pull, dossier push |
| Know who you are, which Workspaces you belong to, which Build you target | Workspace sharing | auth status, build list, build info |
Reads are free and immediate. Audit runs require --confirm. Targeted Finding and Control checks show their price before starting without a confirmation prompt. They can consume Workspace credits. See Pricing and credits.
What this preview does not cover
- The Trust Page, billing, and Build settings stay in the app. So does installing the SDK, which Install the SDK explains, because the CLI reads production evidence and never writes the SDK into an app.
- Windows and Linux on ARM have no release binary.
- The CLI changes no code. You fix in the checkout you already have, with your own Git and your own pull request, and the CLI supplies the evidence beforehand and the recheck afterwards.
Start here
Quickstart
Install, sign in, and read your first Audit. Five minutes, nothing charged.
Coding agent workflows
Give Claude Code or Codex verified Build context before it changes code.
Product model
Build, Workspace, Audit, Control, Finding, Dossier: the words the CLI speaks.
Agents and automation
Stable JSON, exit codes, request IDs, GitHub Actions.
Agent-readable sources
Code agents can start from /llms.txt, read the whole corpus at /llms-full.txt, fetch any page as Markdown, and validate flags against /command-reference.json.
Seven portable skills install for every coding agent on your machine, each depending only on bworlds and these pages:
bworlds-cli: the command contract every other skill loads.bworlds-brief: free orientation on a Build you have just been given access to.bworlds-triage: trace one production error to the code path that produces it.bworlds-fix: take one Finding to a change your own coding agent wrote, then have BWorlds recheck it.bworlds-audit: audit a Build end to end.bworlds-check: a recurring health check.bworlds-guardrails: print a Build's guardrails in the form your own agent reads.
The executable carries them, so one command installs all seven into every coding agent it finds:
bworlds agent setupFor an agent it does not know, the cross-agent installer takes the same skills from this site:
npx skills add https://docs.bworlds.co/bworlds-cli-SKILL.md -g
npx skills add https://docs.bworlds.co/bworlds-brief-SKILL.md -g
npx skills add https://docs.bworlds.co/bworlds-triage-SKILL.md -g
npx skills add https://docs.bworlds.co/bworlds-fix-SKILL.md -g
npx skills add https://docs.bworlds.co/bworlds-audit-SKILL.md -g
npx skills add https://docs.bworlds.co/bworlds-check-SKILL.md -g
npx skills add https://docs.bworlds.co/bworlds-guardrails-SKILL.md -g