17 Nov 2026 · 10:44 AM MTUpdated…
Work, health and model allowance

Overview

PLAYERWEEK / LIVE OPERATIONS

Your system,
at a glance.

What is moving. What is proven. What needs work.

0recent task reports1 area needs a closer look below
LANGUAGE BACKLOG

Capture and processing

Separate counts for news text, podcast audio and local-model work.

14,054news pages without text (including unavailable)
10,440Q-01 pending: 9,784 news for local reader; 656 presser chunks for review or capped Haiku. 4,842 rejected claim rows remain retryable. Reader idle; checkpoint Sep 30 · 11:33 AM MT.
0podcast episodes left to capture
20podcasts left to transcribe
Full progress and schedule ↗
IN MOTION

What is being built

All workers & schedules ↗

Worker task reports expire after 35 minutes. They describe claimed work; a report alone does not prove a process is still running.

No current task reports

Scheduled jobs may be between runs. Check the workforce page for measured process state.

MEASURED HEALTH

What works — and what needs attention

Operational evidence ↗
CAPACITY

How much usage is left

Where the work runs ↗

Allowance and pace come from provider account readings. Current pace is the observed weekly quota change over up to 24 hours; sustainable pace spreads the remaining allowance to reset. Scheduled starts and Codex model runs are separate local activity counts. No model calls are made to refresh this page.

Claude

Weekly32% remaining

Read Nov 17 · 10:35 AM MT

Resets Nov 19 · 7:59 PM MT

Weekly chart
68% usedeven pace 66% at 10:35 AM · over pace
100%75%50%25%0%
No reading Thu Nov 12 · 7 PM MT to Thu Nov 12 · 8 PM MT
FriSatSunMonTueWedThu
Reset Thu Nov 12 · 7:59 PM MTNext reset Thu Nov 19 · 7:59 PM MT

Recorded useEven pace to 100% at resetNo reading

507 readings · 110 of 111 hours so far have a reading. Times Mountain.

Reset in 2.4 days (57.3 hours)

Current pace: 9.08% / day over 23.8 hours

Sustainable pace: 13.41% / day through reset

Logged scheduled starts, past 24 hours: 0

Account-wide allowance; includes work outside Playerweek.

Codex

Weekly16% remaining

Read Nov 17 · 10:43 AM MT

Resets Nov 22 · 3:51 PM MT

Weekly chart
84% usedeven pace 26% at 10:43 AM · over pace
100%75%50%25%0%
No reading Sun Nov 15 · 3 PM MT to Sun Nov 15 · 4 PM MT
MonTueWedThuFriSatSun
Reset Sun Nov 15 · 3:51 PM MTNext reset Sun Nov 22 · 3:51 PM MT

Recorded useEven pace to 100% at resetNo reading

749 readings · 42 of 43 hours so far have a reading. Times Mountain.

Reset in 5.2 days (125.1 hours)

Current pace: 52.05% / day over 24 hours

Sustainable pace: 3.07% / day through reset

Logged scheduled starts, past 24 hours: 15

gpt-6-astra: 17 completed CLI runs in 24 hours · 70,805,812 recorded tokens across 17 runs (4,165,048 per run on average)

gpt-6-luna: 15 completed CLI runs in 24 hours · 571,088 recorded tokens across 15 runs (38,073 per run on average)

gpt-6-sol: 2255 completed CLI runs in 24 hours · 303,125,486 recorded tokens across 2255 runs (134,424 per run on average)

gpt-6.1-sol: 14 completed CLI runs in 24 hours · 149,841,350 recorded tokens across 14 runs (10,702,954 per run on average)

Run model and tokens come from Codex session metadata; account quota is shared across models. Per-model cost and quota draw are not reported.

Account-wide Codex allowance. Direct account-status read; no model request.

Antigravity · Gemini models

Weekly0% remaining

Read Nov 17 · 10:43 AM MT

Resets Nov 20 · 5:23 PM MT

Weekly chart
100% usedeven pace 53% at 10:43 AM · over pace
100%75%50%25%0%
No reading Fri Nov 13 · 5 PM MT to Fri Nov 13 · 6 PM MT
SatSunMonTueWedThuFri
Reset Fri Nov 13 · 5:23 PM MTNext reset Fri Nov 20 · 5:23 PM MT

Recorded useEven pace to 100% at resetNo reading

1,486 readings · 89 of 90 hours so far have a reading. Times Mountain.

Reset in 3.3 days (78.7 hours)

Current pace: 0% / day over 24 hours

Sustainable pace: 0% / day through reset

Logged scheduled starts, past 24 hours: 0

Shared by Gemini Flash and Pro in Antigravity. Local scheduled starts also have a 15-minute cooldown. Gemini app/web usage is unknown here.

Antigravity · Claude and GPT models

Weekly0% remaining

Read Nov 17 · 10:43 AM MT

Resets Nov 21 · 4:34 AM MT

Weekly chart
100% usedeven pace 47% at 10:43 AM · over pace
100%75%50%25%0%
No reading Sat Nov 14 · 4 AM MT to Sat Nov 14 · 5 AM MT
SunMonTueWedThuFriSat
Reset Sat Nov 14 · 4:34 AM MTNext reset Sat Nov 21 · 4:34 AM MT

Recorded useEven pace to 100% at resetNo reading

1,313 readings · 78 of 79 hours so far have a reading. Times Mountain.

Reset in 3.7 days (89.8 hours)

Current pace: 0% / day over 24 hours

Sustainable pace: 0% / day through reset

Scheduled starts: no local ledger for this allowance

Shared by Claude Opus, Claude Sonnet and GPT-OSS in Antigravity; separate from Claude Code and Codex.

UP NEXT

The next work in the queue

Full plan & evidence ↗

Recorded priority order · 200 queued. Assignment can change as workers finish. Board states are claims, not deployment proof.

G-00G-00 — Game-model recipe type and served-pointer switch-over in the release loop (team-strength project, first unit)+

docs/research/TEAM-STRENGTH-PROJECT.md milestone 4 needs a game-model candidate to reach picks, pick'em and the site automatically, and it cannot today. Checked 2026-10-06: playerweek/auto_release.py recognises exactly two recipe schemas, release-candidate/1 (player model) and kernel-candidate/1 (Track B world-model kernel, RESEARCH-CHARTER.md §14, wired by commit 16f4a40f9). Neither is a game-model (team-layer margin/total) recipe. bin/freeze-games.py:289 calls `teamstate.predict(games, teamstate.PUBLISHED)` directly — a hardcoded module constant, not a release pointer — so there is no mechanism by which a better team-strength candidate could ever become served. The shared game-model harness this recipe gates on already exists and is approved: research/scientist/harness/game_pit.py (built by commit ec95084e4, RESEARCH-CHARTER.md §6a) plus research/betting/money_backtest.py and bin/headline-metrics.py (playerweek/exprecord.py APPROVED_HARNESSES). Only the release-loop wiring is missing.

Owner: —

G-01G-01 — Team-strength milestone 1, unit 1 of 4: point-in-time efficiency data assembly+

CLAUDE.md 2026-10-06 "THE FOCUS": we pick 64.1% of winners against Vegas's 66.3%; the ceiling study found the gap lives in team strength. H-377 (CLAIMED): D.ppd carries 89% of the margin residual and EPA/play 80% (research/scientist/RESIDUAL-MAP.md); the gap grew to 6.3, 4.4, 2.9 winners/100 in 2023-25. docs/research/TEAM-STRENGTH-PROJECT.md milestone 1 names the inputs: EPA per play, success rate, points per drive, explosives, turnovers treated as mostly noise, separate pass and run, 2016-2026.

Owner: —

G-02G-02 — Team-strength milestone 1, unit 2 of 4: opponent-adjusted rating fit+

playerweek/teamstate.py's own docstring: "S evidence only: completed scores, identities, season/week and fixtures" — v0.1 learns only from final scores. docs/research/TEAM-STRENGTH-PROJECT.md milestone 1: opponent-adjusted offense/defense ratings from play-level data, separate pass and run, shrunk toward the prior. Depends on G-01's table.

Owner: —

G-03G-03 — Team-strength milestone 1, unit 3 of 4: walk-forward evaluation on the shared harness+

docs/research/TEAM-STRENGTH-PROJECT.md: "Standalone test: winner %, margin and total error vs v0.1 and vs Vegas, by season stage." CLAUDE.md: "Every milestone records winner % vs Vegas on the same games, overall and for weeks 10-18, with n and an interval." Depends on G-02's candidate.

Owner: —

Recent code changes Commits are not deployment receipts
a3f770f94

INC-5e62b18a8f retain build freshness and deploy failure evidence

46 minutes ago
63f9c6cf5

H-521: served kernel kb820d6a still draws nullifying fouls flat by down (R3 0.986 vs real 1.78-2

3 hours ago
68e46415b

residual-ledger: 0 added, board refreshed

4 hours ago
1837bd804

INC-56c2ed4c84: remeasure ledger staleness (player 78.5%->9.9%, max 7.06->8.19); baselines tight

5 hours ago
2849fc86d

Publish 2026-11-17-astra weekly review

6 hours ago
e9360bff3

Publish verify: retry independent edge read for propagation lag (INC-0b062efa15)

6 hours ago
4d2a9e3ab

H-687: scan survivor points_l3_diff is absorbed by as-of team latents (calibration)

6 hours ago
8fb7201ec

Record H-703 source refusal in research ledger

8 hours ago
85d6691a6

Record H-703 scoring-channel source census

8 hours ago
569a1674c

ceo-periodic-queue: 6 item(s) queued for 2026-11-17

10 hours ago

Refreshes automatically after the scheduled collection and publication, normally every five minutes. Source readings have their own timestamps. Missing or stale evidence is never a pass.