← charleslikesdata.com afk cockpit
9 repos · 996 runs · 816 issues · 725 merged · data through Jun 7 – Aug 3, 2026

Autonomous-PR outcomes across the fleet. Cost is API-equivalent (notional on a Pro subscription). Quarantines are a safety signal — afk refused to ship — not failures.

Fleet state · not yet wired up
Live gate/promote/merge and dependency-frontier state isn't ingested yet — this panel is a placeholder for a future feature, not a fabricated number.
98%
landed rate
n = 739 gate-passed
98%
landed rate (executor repos)
n = 8 exec repos
excludes breakroom (0 quar)
$3.90
API-equiv $ / merged PR
n = 725 merged
fully loaded: includes failed runs, quarantines, and scout overhead
33%
quarantine (safety) rate
n = 816 issues
18%
escalation rate
n = 996 runs
0%
rejected rate
n = 1,423 proposals
Pipeline

Stage funnel

proposed
1,423
promoted
840
59% of proposed
attempted
816
97% of promoted
passed gate
739
91% of attempted
published
728
99% of passed gate
merged
725
100% of published
By repository

Outcomes per repo

landed quarantined other
Orbit-Wars
77 · 87%
afk-agent-system
431 · 90%
afk-cockpit
106 · 94%
breakroom
2 · 0%
graphmark
14 · 93%
household-bms
48 · 96%
household-command-center
1 · 100%
housing-commute-analysis
15 · 93%
the-workshop
122 · 80%
Producers

Proposal quality

dropped
216
harvest
4
mis-sized
75
pending
286
rejected
1
shipped
840
superseded
1
Safety axis

Quarantine by category

271 total
other
84
no-op
35
review
24
capability
21
integration-conflict
17
build-crash
16
scope
14
setup-crash
10
test-conflict
10
deny-surface
9
circuit-breaker
7
timeout
5
coupling
4
idle-timeout
4
environmental
3
unrecoverable-signature
3
flaky-test
2
security
2
rate-limit
1
Retries & recovery

Runs per issue

1.22 avg runs/issue
12345
Retries & recovery

Quarantine → retry rescue

retry rescues 62% of first-run quarantines · 106% of those actually retried
75%
first-try pass rate
16%
final-state quarantine (issues)
27%
quarantine rate (runs)
Cost · API-equivalent
fleet telemetry · per attempt · Jun 7 – Aug 3, 2026

Spend by model tier

$2,511.38 total across the window
sonnet$1628.92 · avg $1.68/attempt · n=968 attempts
unknown tier$670.16 · avg $1.69/attempt · n=397 attempts
opus$212.29 · avg $2.47/attempt · n=86 attempts
⚠ 27% of attempt spend ($670.16 · 397 attempts) has no recorded model_tier — telemetry gap, not a model
Overhead · orchestrator + scout ops
$318.14
286 cost events · additive to the tier spend above · not a model tier
Claude Code transcripts, afk worktrees only · per API call · Jul 4 – Aug 3, 2026
$0.03
median / call
$0.08
p90 / call
22,200
calls logged
whole machine, all projects: $8,921.03 across 60,777 calls
Cost dial

Per-API-call cost distribution

16764
4394
748
185
85
19
5
<$.05.05–.10.10–.15.15–.20.20–.30.30–.50$.50+
Model mix

Opus share by day

Jul 4 range: 0%–74% Aug 3
Throughput

Slices per day

996 runs
06-0706-1306-1906-2707-0307-1407-2307-30
Context cost

Tokens by type, per day

1.59B tokens · 95.6% cache reads — the context is overwhelmingly cached
cache read cache write input output