Manage Railcode Org
Update First
Before answering a Railcode management or CLI question or running a
command,
update the installed Railcode skills and CLI, then confirm npm's published version:
bash
npx skills add Railcode-HQ/railcode-skills
npm install -g railcode@latest
railcode --version
npm view railcode version
If the skill changes, re-read this file from the top. If npm is unreachable, state that the
latest version could not be verified and do not claim this guidance is current. This version
was checked against published Railcode CLI 0.1.28.
Since 0.1.28 the CLI self-updates within its major version — but only on an
interactive
terminal, and agent-driven sessions are non-interactive, so keep running the explicit
npm install -g railcode@latest
above rather than assuming you're on the latest.
Management Workflow
1. Authenticate and establish scope
Run
if needed. Management commands target the organization saved by login
and work from any directory; they do not require an app or
.
Before mutating state, confirm the intended organization and inspect the current resource.
Most references accept a name, slug, or email as appropriate, or a UUID. Prefer UUIDs when a
human-readable reference is ambiguous. Use
when a machine-readable result matters.
2. Inspect before changing
Use the matching read command first:
,
,
roles list/grants/effective/catalog
,
,
/
,
,
, or
.
Owner/admin permissions are enforced server-side. A
is an authority boundary, not a
reason to bypass the CLI or use another credential without the user's authorization.
3. Apply the narrowest mutation
Change only the named resource. Preserve least privilege:
- grant specific resources instead of unless broad access is explicitly intended;
- prefer saved queries over ad-hoc SQL authority;
- restrict service-connector HTTP methods;
- avoid embedding credentials in shell history—prefer the supported file options;
- do not delete, remove a member, transfer ownership, revoke access, or rotate credentials
unless the user requested that state change.
4. Verify effective state
Repeat the relevant read command after mutation. For access changes, verify both the direct
policy and computed grants (
,
). For connector setup, validate
with the least invasive list/docs/query operation that proves configuration without causing
unrequested downstream side effects.
Capability Boundaries
- is readable by any member; member mutations require admin authority.
- App list/show/access follow per-app visibility; set-access/transfer/archive/unarchive/delete
require manage rights (owner or org admin).
- Roles, grants, connections, connector administration, analytics, and logs are capability-
gated server-side.
- The CLI cannot create a member. A new person joins through the invite flow in the web app;
the CLI then lists, re-roles, or removes them.
- removes deploys and app data and is irreversible. is the
reversible alternative — the app keeps serving, keeps its data, and keeps running its
agents; it only leaves the launcher. Propose archiving whenever the user's goal is to
retire an app rather than destroy it.
- / (singular ) read and write a deployed app's records and files.
They need an owner grant or , and /// mutate live
tenant data — including individual members' private scopes. Inspect first; mutate only what
was asked for.
- expands a wildcard into explicit rows; inspect the subject and resource
first because it changes future grant maintenance semantics.
Skill Boundaries
Use
for scaffolding, developing, testing, manifesting, and deploying a
static app. Use
for managed-agent JSON manifests, tests, invocations,
and schedules. This skill owns the organization-level administration those builders may
depend on.
Reference
Read CLI management reference for exact commands, flags,
resource types, access modes, connection shapes, log filters, and saved-query administration.