Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/xantorres/repokernel/llms.txt

Use this file to discover all available pages before exploring further.

rk team status is RepoKernel’s answer to “what is each agent doing right now?” — a single command that joins runtime data from run files, declared state from the registry, and operational lane state into one coherent snapshot. Instead of grep-ing log files across four terminal tabs, you get runs, sprints, registry health, and current bottlenecks in one place.

​
What the dashboard shows

Running rk team status produces output like this: 
team status — 2026-05-04T13:30:00Z

Runs
RUN     EPIC      STATUS    ACTIVE  READY  REVIEW  STARTED              ETA
RUN-005 E-team-core running 3       1      1       2026-05-04 13:00:00  2026-05-04 14:12:00
RUN-006 E-042     running   1       0      0       2026-05-04 13:25:00  2026-05-04 14:01:00

Sprints
SPRINT          STATUS  LANE  AGENT  PROGRESS  TITLE
S-merge-safe    active  core  claude 67%       P0: Merge-Safe State
S-team          active  core  codex  33%       P1: Team Status Visibility

Registry
  health=OK  ready_to_merge=true  conflicts=0  files_changed=0

Bottlenecks
  • S-tracker: awaiting_review

​
Runs

Each row in the Runs section represents one active run:
ColumnMeaning
STATUSrunning / paused / completed / failed / aborted
ACTIVENumber of sprints currently in flight for this run
READYSprints in the queue, eligible to dispatch
REVIEWSprints awaiting a review verdict
STARTEDWall-clock start time
ETADerived from elapsed time, completed-sprint count, and remaining work

​
Sprints

The Sprints section shows a per-sprint snapshot. The AGENT column resolves to whichever run owns the active claim. PROGRESS is exact when run.limit is set, and shows a raw sprint count otherwise.

​
Registry health

FieldMeaning
healthOK / DEGRADED (P1 findings present) / BLOCKED (P0 or threshold-meeting findings)
ready_to_mergetrue only when health is OK
conflictsCount of P0 findings (true blockers)
files_changedCount of files carrying findings across P0+P1+P2 — surfaces triage volume

​
Bottlenecks

The Bottlenecks section lists conditions that are actively blocking forward progress:
  • S-NNN: awaiting_review — review verdict outstanding
  • S-NNN: blocked_by S-MMM — declared dependency unmet
  • E-NNN: K concurrent runs (lane saturated) — multiple runs in flight on the same epic

​
Flags

rk team status                        # human-friendly dashboard
rk team status --json                 # machine-readable; pipe into anything
rk team status --watch                # refresh every 30s (15s minimum floor), SIGINT-safe
rk team status --watch --interval 60  # custom refresh interval (seconds)
rk team status --sprint S-042         # drill into a single sprint
Use --watch during long multi-agent runs for a live refreshing terminal view. The loop clears the screen between iterations and exits cleanly on SIGINT or SIGTERM — no stray listeners left behind.

​
JSON output

--json emits a stable, Zod-validated envelope (TeamStatusSchema in @repokernel/core) suited for dashboards and agent chaining: 
rk team status --json | jq '.runs[] | {id: .run_id, status: .status, eta: .eta}'

{
  "timestamp": "2026-05-04T13:30:00.000Z",
  "runs": [
    {
      "run_id": "RUN-005",
      "epic_id": "E-team-core",
      "status": "running",
      "active_sprints": 3,
      "states": { "ready": 1, "active": 1, "review": 1, "merging": 0 },
      "started_at": "2026-05-04T13:00:00.000Z",
      "ended_at": null,
      "eta": "2026-05-04T14:12:00.000Z"
    }
  ],
  "sprints": [
    {
      "id": "S-merge-safe",
      "title": "P0: Merge-Safe State",
      "status": "active",
      "agent": "claude",
      "lane": "core",
      "run_id": "RUN-005",
      "progress": "67%",
      "started_at": "2026-05-04T13:00:00.000Z",
      "eta": "2026-05-04T14:12:00.000Z"
    }
  ],
  "registry": {
    "files_changed": 0,
    "conflicts": 0,
    "ready_to_merge": true,
    "health": "OK"
  },
  "bottlenecks": ["S-tracker: awaiting_review"]
}

​
Project-level views

Beyond rk team status, RepoKernel ships several commands for per-entity inspection.

​
rk report

rk report prints a lean project snapshot directly to your terminal: a headline, the next runnable sprint, sprint groups by status, and active findings. 
rk report          # headline, next sprint, sprint groups, findings
rk report --all    # include shipped/done sprints and the full epic table
rk report --json   # full structured report for machine consumption

​
Per-entity list commands

For focused inspection, use these targeted list commands: 
rk ls epics           # list all epics
rk ls sprints         # list all sprints
rk lane ls            # list lanes and their current state
All list commands accept --json for machine-readable output.

​
Data sources

The dashboard composes data from three independent sources:
  1. Run files (live) — the active run records under <git-common-dir>/repokernel/
  2. Registry (declared state) — .repokernel/registry.json, the merged view of all sprint and epic state
  3. Operational lane state — per-lane claim and queue files
This means rk team status gives you an accurate picture even when multiple agents are running simultaneously on different branches.

Build docs developers (and LLMs) love

Get started for freeTalk to us