| name | box-status |
| description | Report the local box-memory workspace state - Box Drive backend health, teams, memory counts by kind/status, recent activity, index health. Read-only. Use when the user asks what's in their workspace, wants a status report, asks if Box Drive is syncing, or wants a sanity check before a heavy operation. |
| argument-hint | [--check-indexes] [--team=<name>] [--no-cache] |
/box-status
If you see unfamiliar placeholders or need to check which tools are connected, see CONNECTORS.md.
Produce a human-readable snapshot of the local box-memory workspace. Read-only — never modifies anything. Useful as pre-flight before heavy operations, an audit summary, or just to remind the user what's there.
Usage
/box-status [--check-indexes] [--team=<name>] [--no-cache]
Examples:
/box-status — standard report/box-status --check-indexes — also flag suspect index drift/box-status --team=engineering — scope to a single team/box-status --no-cache — re-run box-drive-detect instead of using cache
What to do
- Read
_box-memory.json. If missing, output: "No box-memory workspace found at this Box Drive mount. Run /box-init to create one." Don't error.
- Backend status. Read cached
box-drive-detect result, or re-probe if --no-cache. Report:
- Box Drive mount path
- Mount writable
- Box Drive process running
- Account email
- Sync state (idle / pending / offline / unknown)
- Workspace contents. Walk per-team
memories/_index.json files:
- Total memories
- Counts by status (active / draft / superseded / archived)
- Counts by kind (decision / fact / task / observation / reference / note / companion)
- Companions count
- Files count (binaries in
files/ and team files/ folders)
- Per-team stats. Scoped if
--team=<name>; else all teams.
- (If
--check-indexes) For each per-folder _index.json, compare its updated_at to the latest file mtime in that folder via local FS. If folder has newer files than the index, surface drift — suggest /box-index-rebuild. Does NOT rebuild.
- Compose report:
box-memory-onprem workspace: <workspace-name>
Workspace path: <absolute local path>
Created: <ISO>
Updated: <ISO>
Backend: local (Box Drive)
Owner app: box-memory-onprem
Plugin version: <version>
Box Drive:
Mount: <path>
Writable: <yes/no>
Process running: <yes/no>
Account: <email or unknown>
Sync state: <state>
Workspace contents:
Total memories: <N>
active: <n>
draft: <n>
superseded: <n>
archived: <n>
By kind:
decision: <n>
fact: <n>
task: <n>
observation: <n>
reference: <n>
note: <n>
companion: <n>
Binary files: <N>
Teams:
default <N> memories, <N> files
<other teams> ...
Index health:
Per-folder indexes: <N>
Latest index update: <ISO>
Drift suspected: <none | list of folders> (only if --check-indexes)
Most recent memories:
- <title> (<kind>, <updated_at>, <team>)
- ... (up to 5)
Top tags:
<tag>: <count>
... (up to 5)
Settings:
Default team: <name>
Companion layout: <sibling | folder>
Include superseded: <yes/no>
Auto-rebuild on drift: <yes/no>
Status as of: <ISO-8601>
- Flag actionable issues:
- Box Drive not running but mount exists → "Files written locally will NOT sync to Box cloud until Box Drive is started."
- Drift suspected from
--check-indexes → → Run /box-index-rebuild <folder>
- Compliance target declared but no LLM declaration →
→ Run /box-airgap-status to verify LLM provider matches your target
- Stale lockfile present → suggest
/box-index-rebuild to clear it
Don't
- Don't modify anything. Status is read-only.
- Don't run
box-drive-detect automatically without --no-cache — the cache is the point.
- Don't rebuild indexes during status. If drift is suspected, surface and suggest
/box-index-rebuild.
- Don't load every memory's full content. Index entries are enough for a status report.
- Don't call any Box API or read any URL. Local FS only.
Errors
- Workspace not initialized → suggest
/box-init; informational, not error
_box-memory.json corrupt / unparseable → surface clearly; suggest /box-index-rebuild --check to assess- Some team folders inaccessible → list which, continue with the rest
