Token Burn

Burn is not proof. Output is not victory. Accepted work that repeats is leverage.

today · in flight
465K active
13 sessions · spend 821K · Claude Code 465K
by project: <home> 465K
top session 54704d28 (<home>, 36K, —)
2026-09-22
as of 2026-09-22T13:00:15Z
Stupid button · 28m ago · gemma4:e4b · 2131t in / 357t out · 49.9s

Trend vs last week

What worked

What wasted tokens

Why the big number is big

No agent fan-out this window.

What needs a verdict

One contrarian observation

The <home> project has the largest waste volume at 5.5M (5.5M in waste_active_h), but this volume is unmeasurable because it ran outside any git repo.

---

Measurability caveat (generated, not model-authored): <home> (5.5M) ran outside any git repo. There is no diff to credit, so a 0.0 acceptance ratio there is the meter failing to see the work — not a judgement on it. Only ratios on measurable: true projects are verdicts.

Rollups — acceptance lane

Headline is accepted tokens (active lane, manual label wins over git auto-classification). Unsettled = committed-not-merged + open-loop. "Active" excludes cache replay; "spend" includes it.

Last 7 days
0 accepted
unsettled 1.6M · acceptance 0%
spend 30.7M · 157 sessions · Claude Code 7.1M
Last 30 days
0 accepted
unsettled 39.5M · acceptance 0%
spend 884.4M · 731 sessions · Claude Code 43.6M / Codex 4.6M
Last 90 days
0 accepted
unsettled 129.8M · acceptance 0%
spend 3.70B · 841 sessions · Codex 87.0M / Claude Code 51.7M

Daily heatmap — last 60 days

Each cell = one UTC day, color intensity = log10(active tokens). Hover for date + counts.

LessMore· log-scale, exact lane
29.1M on 07-28

Top 10 days

DateActiveSpend SessionsTop projectOutcomes
2026-07-2829.1M1.09B22<home> (29.1M)?=22
2026-08-1821.2M527.5M12<home> (21.0M)?=14
2026-07-1615.7M522.1M23<home> (15.7M)?=23
2026-08-0611.1M362.9M27<home> (11.1M)?=27
2026-08-315.9M144.8M26las (5.8M)?=29
2026-09-085.1M47.8M28las (4.0M)?=28
2026-08-294.5M158.4M30las (1.7M)?=31
2026-08-304.2M72.4M25las (4.2M)?=25
2026-09-044.1M90.1M29las (2.3M)?=33
2026-09-113.9M69.9M30las (3.0M)no_repo=24 · open_loop=5 · committed=1 · PRs #21

By project — last 7 days

ProjectActive 
<home>5.5M
las1.6M

By outcome — last 7 days

committed 1 #21 (OPEN)
no_repo 156

Effectiveness — last 30 days

Where delegated work shipped (top) vs. where high burn never crossed an acceptance line (bottom). Bars: accepted · unsettled · waste. Normalized to the widest project in window.

Leverage — most accepted

ProjectAcceptedRatioMix
las00%
pal00%
memory00%
mainstay-av00%
pal/templari00%
Downloads00%

Waste candidates — lowest acceptance ratio (≥100K active)

ProjectAcceptedRatioMix
las00%
<home>00%
mainstay-av00%
pal00%
memory00%

Needs validation — last 14 days (8 unsettled)

Sessions where auto-classification couldn't tell shipped work from waste. Highest-burn first. Manually label these with `label: "used"|"partial"|"redo"|"review"|"not_useful"` in token_burn.jsonl to override the auto verdict.

DateProjectSessionActiveAutocwd
2026-09-11las4b15a9973.0Mcommitted · #21 (OPEN).../las/.worktrees/mvp
2026-09-09las4b15a9972.7Mcommitted · #21 (OPEN).../las/.worktrees/mvp
2026-09-16las4b15a9971.6Mcommitted · #21 (OPEN).../las/.worktrees/mvp
2026-09-11pal01a07347448Kopen_loop.../pal
2026-09-11pal01a0925e165Kopen_loop.../pal
2026-09-11pal01a0925f151Kopen_loop.../pal
2026-09-11pal01a0925e17Kopen_loop.../pal
2026-09-11pal01a0888c17Kopen_loop.../pal

Recent events — last 30 rows

Latest enriched rows from the JSONL (sorted by date desc, active desc as tiebreaker). M = manually labeled.

DateToolProjectSessionActiveOutputOutcome
2026-09-22CC<home>54704d2836K928no_repo
2026-09-22CC<home>7670347a36K826no_repo
2026-09-22CC<home>c1c2e6f836K724no_repo
2026-09-22CC<home>e5f1a06f36K716no_repo
2026-09-22CC<home>4aaec21536K994no_repo
2026-09-22CC<home>c5b9885736K748no_repo
2026-09-22CC<home>73a85ea636K738no_repo
2026-09-22CC<home>3bcb601c36K606no_repo
2026-09-22CC<home>7be5c0d736K590no_repo
2026-09-22CC<home>a958014836K566no_repo
2026-09-22CC<home>111a17b536K702no_repo
2026-09-22CC<home>9ac9283d36K664no_repo
2026-09-21CC<home>6d72afa636K1Kno_repo
2026-09-21CC<home>5cbd834536K1Kno_repo
2026-09-21CC<home>70d8128536K902no_repo
2026-09-21CC<home>d836925336K894no_repo
2026-09-21CC<home>ddfb634536K890no_repo
2026-09-21CC<home>d9bb83b436K818no_repo
2026-09-21CC<home>502176f436K768no_repo
2026-09-21CC<home>09a8bc5b36K750no_repo
2026-09-21CC<home>2b94da7b36K706no_repo
2026-09-21CC<home>62d349c036K702no_repo
2026-09-21CC<home>33d5061736K704no_repo
2026-09-21CC<home>607944ba35K640no_repo
2026-09-21CC<home>c12c593235K628no_repo
2026-09-21CC<home>561cb96d35K612no_repo
2026-09-21CC<home>9e8d927435K612no_repo
2026-09-21CC<home>4d5740fa35K606no_repo
2026-09-21CC<home>92d03b1035K432no_repo
2026-09-21CC<home>c0c4eef735K594no_repo

Generated 2026-09-22 13:00 UTC from ~/ideas/pal/dashboards/token_burn.jsonl (909 rows). Sources: ~/.codex/sessions/ (exact), ~/.claude/projects/-home-khoward/ (exact). No estimate lane currently active.