Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135

Activity

1. developing-today commented on Oct 29, 2026

   

   developing-todayBot

   on Oct 29, 2026 – with developing.today

   Contributor

   More actions

   CODE-1033
2. dezren39

   added

   priority/highImportant and time-sensitiveImportant and time-sensitive

   on Oct 29, 2026
3. dezren39 commented on Oct 29, 2026

   

   dezren39

   on Oct 29, 2026

   MemberAuthor

   More actions

   TOP PRIORITY (Drew, 2026-10-29 01:23 CDT): Batch 3 fleet consolidation — pick this up in the next runs, ahead of routine work. Being enqueued to repo-health-code-pilot at top priority now.
4. dezren39

   added

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   work/assignedQueue worker lifecycle stateQueue worker lifecycle state

   queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

   status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

   and removed

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   on Oct 29, 2026
5. dezren39 commented on Oct 29, 2026

   

   dezren39

   on Oct 29, 2026

   MemberAuthor

   More actions

   🤖 Agent note · picked up

   • Agent: queue worker (repo-worker-code), lane D
   • Session / ID: queue task 3165
   • Found via: queue claim
   • Task: Triage Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135 (pool-quota-steward snapshot-first change; fleet efficiency batch 3, epic epic: fleet expansion & placement program (2026-10-28 directive, LIST 1–5) #2055): verify premise against live state, produce the design/decision, route implementation in the same pass, update epic epic: fleet expansion & placement program (2026-10-28 directive, LIST 1–5) #2055.
   • Expectation: done = verdict comment on Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135, follow-on packet enqueued, epic comment posted, labels truthful.
   • Links: epic epic: fleet expansion & placement program (2026-10-28 directive, LIST 1–5) #2055
6. dezren39 commented on Oct 29, 2026

   

   dezren39

   on Oct 29, 2026

   MemberAuthor

   More actions

   Triage verdict — task 3165 (lane D)

   Premise: VERIFIED against live state (structure and behaviour; the token figures are the batch's own measurement, not re-measured here). Decision: implement snapshot-first exactly as scoped below. Implementation routed in this same pass: follow-on queue task 3177. Epic #2055 updated.

   Premise check

   • The steward re-probes everything, every run. Its own ledger (pool-change-ledger.md, runs 2026-10-29T05:24Z and 06:24Z) records, per run, two independent fresh laptop /api/quota/* reads (identical used% — the fleet's verify-twice habit) spanning meta, opencode-go, cline-pass, codex, openai.alternate, openrouter, google (20 models), github-copilot and deepseek, plus five failing read attempts (command-code, claude, antigravity, opendesign, openai endpoint), drift re-checks, a dispatcher-log note, and impact follow-ups measured for prior events — every hour. The job itself is live and hourly (cron pool-quota-steward, interval 1h, enabled).
   • The compiled snapshot exists and is fresher than the gate needs. The aggregator's gatherer (pm-stack/aggregator/scripts/gather-quota.ts) ticks every 15 minutes and writes quota-current.json — per-provider windows with measuredAt, state, source — ingesting the OpenChamber quota module plus file-export producers (Meta's ~/atlas/meta-quota-export/latest.json, verified present on the laptop today). provider-quota-registry.md is maintained alongside. So for any pool the hub reads successfully, a ≤15-minute-old compiled figure already exists when the steward runs.
   • The gate constraint is real and is the design's centre. The charter measurement gate forbids a placement without a quota read ≤60 min old for that pool. Snapshot-first is safe for that gate only because the hub tick (15 min) is fresher than the gate (60 min) — and only for pools whose hub entry is actually fresh and ok.
   • The token figures (3.45M input/run, 51.8M per 15h, target ≤1.7M) come from the fleet-efficiency batch's check-in, cited on this issue and epic epic: fleet expansion & placement program (2026-10-28 directive, LIST 1–5) #2055; this triage did not re-measure them. The follow-on packet therefore includes a before/after measurement (3 runs each side) recorded in the ledger, so the saving is proven, not assumed.

   Decision / design (the routing in task 3177)

   1. Snapshot-first: a steward run consumes any pool whose hub entry (quota-current.json) has measuredAt ≤60 min and state: ok. No live probe for those pools.
   2. Probe only what's stale, absent, or broken — plus pools the dispatcher is actively drawing from (seated in live sessions), which keep hourly live probing. Today the stale/broken set is typically command-code (read broken, feat(quota-plane): adapter — Command Code via CLI/app credential #1165), claude, antigravity, opendesign — naming them in the run report stays mandatory, so coverage does not silently shrink.
   3. Follow-ups daily: impact verdicts are evaluated in the first steward run of each day (America/Chicago); other runs append a one-line follow-ups due: N, next daily pass <date> ledger note, so nothing sits silently — the daily pass is a named owner, not a drop.
   4. Not changed, explicitly: the hourly cadence (a rate change would need Drew's discussion under the 2026-10-28 efficiency rule — this is a method change, pre-approved class), the gate, the broken provider reads themselves (feat(quota-plane): adapter — Command Code via CLI/app credential #1165, opencode#818 own those).

   Routing

   Follow-on packet enqueued: queue task 3177 (pool-quota-steward: snapshot-first quota picture… — fix #2135, packet hidden_files/packet-3165-fix.json), acceptance covering the instruction + job-body update, the first post-change ledger entry showing the consumed-vs-probed split per pool, and the before/after token measurement.

   🤖 Agent note · done

   • Agent: queue worker (repo-worker-code), lane D
   • Session / ID: queue task 3165 (triage); follow-on task 3177
   • Found via: queue claim
   • Task: Triage Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135 (fleet efficiency batch 3) — premise verified, design decided, implementation routed, epic updated.
   • Expectation: done as stated above; Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135 stays open until task 3177 lands and its first post-change run evidences the split.
   • Links: epic epic: fleet expansion & placement program (2026-10-28 directive, LIST 1–5) #2055 · follow-on task 3177 · broken-read owner feat(quota-plane): adapter — Command Code via CLI/app credential #1165
7. dezren39

   added

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   and removed

   status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

   queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

   work/assignedQueue worker lifecycle stateQueue worker lifecycle state

   on Oct 29, 2026
8. github-actions

   assigned

   dezren39

   on Oct 29, 2026
9. 37 remaining items

   Load more
10. dezren39

    added

    source/creatorFiled by the new-issue-creator jobFiled by the new-issue-creator job

    status/needs-triageNot yet triaged; needs a first lookNot yet triaged; needs a first look

    task/triageQueue packet kind: verify, reproduce, and route this itemQueue packet kind: verify, reproduce, and route this item

    on Nov 3, 2026
11. developing-today commented on Nov 3, 2026

    

    developing-todayBot

    on Nov 3, 2026 – with developing.today

    Contributor

    More actions

    agent-note:v1
    state: reviewed (coverage loop-back)
    Coverage pass 2026-11-03, loop-back review: this issue was re-reviewed (drift / overlap / already-handled / stale / blockers / ordering) so every open issue in the four repos is reviewed and in the system. Verdict: still live; in the system as task 7298 (triage, queued, priority 50) in repo-health-code-pilot. Labels audited and synced. If it is already handled at current main, its triage closes it with evidence; if oversized, the triage files the split; blockers belong on this issue.
    Found via: coverage loop-back 2026-11-03
12. dezren39

    added

    work/assignedQueue worker lifecycle stateQueue worker lifecycle state

    queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

    status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

    and removed

    status/needs-triageNot yet triaged; needs a first lookNot yet triaged; needs a first look

    queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

    on Nov 3, 2026
13. dezren39 commented on Nov 3, 2026

    

    dezren39

    on Nov 3, 2026

    MemberAuthor

    More actions

    🤖 Agent note · picked-up

    • Agent: queue worker (repo-worker-code), lane code-m
    • Session / ID: task 7298
    • Found via: queue claim
    • Task: task 7298 triage of issue Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135 (pool quota steward: consume compiled hub numbers, probe only stale pools, follow-ups daily). Packet belief: already implemented — verifying against the live steward instruction and today's ledger before closing.
    • Expectation: close with evidence if all three changes verify; otherwise name the gap and enqueue a fix path.
    • Links: issue Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135
14. dezren39 commented on Nov 3, 2026

    

    dezren39

    on Nov 3, 2026

    MemberAuthor

    More actions

    🤖 Agent note · done — closing with evidence (already implemented)

    • Agent: queue worker (repo-worker-code), lane code-m
    • Session / ID: task 7298
    • Found via: queue claim
    • Task: task 7298 triage of issue Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135 verified at 2026-11-03T23:37Z — all three changes in this issue are implemented and live:
      1. Snapshot-first: pool-quota-steward-instruction.md step 1 ("snapshot-first (Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135, 2026-10-29)") starts from the aggregator hub's compiled quota-current.json + the quota registry + the last ledger snapshot, and consumes without probing any pool whose hub entry is fresh (measuredAt ≤60 min, state ok), recorded as consumed-from-hub.
      2. Probe only stale: live probing is limited to pools that are stale (>60 min), absent, or in an error state in the hub picture, plus pools actively drawn from in live sessions — so the ≤60-min measurement gate stays honest for active pools.
      3. Follow-ups daily: step 5 ("Impact follow-up — daily, not hourly (Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135, 2026-10-29)") evaluates impact follow-ups in one designated run per day (first steward run of the day, America/Chicago); other runs append a follow-ups due: N, next daily pass <date> note.
    • Evidence: the implementation change event is in pool-change-ledger.md (2026-10-29: "step 1 rewritten snapshot-first … Step 5 impact follow-ups move to ONE designated daily run … cadence unchanged (hourly)"), and today's steward runs (e.g. the 2026-11-03T22:45Z hub snapshot entry) record "consumed-from-hub where fresh + state ok, live-probed per step 1(d) for actively drawn pools and the error pool" — the behavior is operating, not just written. The hourly pool-quota-steward cron body itself cites Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135 for exactly this behavior.
    • Expectation: issue closed as completed/already-done; no fix path needed. The ≤1.7M input/run figure was the design target for this change; this triage verified the structural change and its live operation, not a fresh token audit.
    • Links: issue Pool quota steward: consume the compiled hub numbers — probe only stale pools, follow-ups daily #2135
15. dezren39

    closed this as completedon Nov 3, 2026
16. dezren39

    removed

    status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

    queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

    work/assignedQueue worker lifecycle stateQueue worker lifecycle state

    on Nov 3, 2026