Orchestration Runs

Account status — using account 1

• Window usage so far: 27,771,773 tokens ($15.93) — 204% of its own last observed ceiling
• Total dispatches on this account: 58

Account status — using account 1

• Window usage so far: 27,811,419 tokens ($16.17) — 204% of its own last observed ceiling
• Total dispatches on this account: 59

Phase Timeline¶

ORCHESTRATOR TOTAL: 0s, 0 tokens, $0.0000 (rows 7)
SHARED TOTAL: 39s, 78,275 tokens, $0.3445, 3 row(s) with no token data (rows 1–6)

Total Run Time: 22m (wall clock, 24-09-2026 19:27:04 → 24-09-2026 19:49:06 IST — not a sum of the per-side totals above, since sides overlap)
Active time: 2m | Waiting (session limits, approvals, dead sessions): 20m — 1 gap(s) still unattributed; label them with driver.py gap <slug> --from <ISO> --to <ISO> --reason '<why>'
Total Tokens Consumed: 78,275 (plus 3 row(s) whose usage was never captured — run execution_tracker.py enrich-audit to backfill. This is a floor, not the total.)
Token breakdown: input 5,126 · output 1,211 · cache-create 50,374 · cache-read 21,564 (28% of the total). Cache-read is context a node re-sends to itself across its own tool-use turns, billed at roughly a tenth of fresh input — a large Tokens number that is mostly cache-read is a long tool loop, not a large workload.
Total Cost: $0.3445 (plus 3 row(s) whose usage was never captured — run execution_tracker.py enrich-audit to backfill. This is a floor, not the total.)

Account status — using account 1

• Window usage so far: 27,850,048 tokens ($16.28) — 205% of its own last observed ceiling
• Total dispatches on this account: 60

Account status — using account 1

• Window usage so far: 28,234,735 tokens ($16.86) — 207% of its own last observed ceiling
• Total dispatches on this account: 61

Account status — using account 1

• Window usage so far: 28,475,380 tokens ($17.20) — 209% of its own last observed ceiling
• Total dispatches on this account: 62