Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Launch a finished Codex Process Jobs record again as a new detached job using its validated persisted argv, working directory, and execution mode. Use only when the user explicitly asks or approves rerunning a specific completed, failed, or cancelled build, test, benchmark, inference run, data job, or repair.
.claude/skills/hashgraph-online-rerun/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-15 | ✗→✓ | ▲ Improved | 144% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-24 | ✗→✓ | ▲ Improved | 586% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -6% | 0% |
Resolve <plugin-root> two directories above this file and run:
textnode "<plugin-root>/scripts/job.mjs" rerun <job-id> [options] --json
Never search memory for CPJ work; use validated CPJ state.
Require a specific source job. If identity is unclear, use $status once to find it. Never reconstruct argv from displayed command text, logs, or model memory. The controller reads the private validated record, refuses active jobs, and creates a fresh job ID and logs with rerunOf lineage.
A rerun repeats the invocation, not the historical environment: files, dependencies, environment variables, credentials, devices, and external state may have changed. Mention a material change before launch; do not claim exact reproducibility.
Never rerun automatically from a completion notice. A direct user request authorizes an ordinary non-critical rerun. For a CRITICAL job, explain that the command may repeat repair, migration, firmware, or destructive effects and obtain explicit risk-aware approval immediately before adding --force.
Add --goal-mode only for an explicitly active Goal. Optional flags are --no-notify, --notify-user, and --no-notify-user. If the durable state directory is not writable in the current sandbox, request the same narrow controller escalation described by $start.
Treat a successful rerun as the same hard release boundary as $start: report the new job ID and its source job ID, state that completion is recorded, then end the turn without status, tail, result, wait, sleep, or process monitoring. If the controller refuses because the source remains active, its working directory disappeared, or legacy shell semantics cannot be preserved, report that refusal and do not improvise a replacement command.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 9,999 | 13,536 | +35% | 1 | 1 | 0% | 264 | 759 | +188% | 0 | 0 | — |
case-02 | fail→fail | 20,021 | 11,250 | -44% | 1 | 1 | 0% | 679 | 725 | +7% | 0 | 0 | — |
case-03 | fail→fail | 11,571 | 13,358 | +15% | 1 | 1 | 0% | 1,924 | 827 | -57% | 0 | 0 | — |
case-04 | pass→pass | 7,974 | 4,410 | -45% | 1 | 1 | 0% | 1,223 | 1,149 | -6% | 0 | 0 | — |
case-05 | fail→fail | 2,647 | 22,594 | +754% | 1 | 1 | 0% | 375 | 2,166 | +478% | 0 | 0 | — |
case-06 | fail→fail | 7,175 | 23,880 | +233% | 1 | 1 | 0% | 1,187 | 2,507 | +111% | 0 | 0 | — |
case-07 | fail→fail | 7,448 | 13,278 | +78% | 1 | 1 | 0% | 335 | 1,052 | +214% | 0 | 0 | — |
case-08 | pass→pass | 8,586 | 5,336 | -38% | 1 | 1 | 0% | 1,395 | 1,256 | -10% | 0 | 0 | — |
case-09 | pass→pass | 20,283 | 10,619 | -48% | 1 | 1 | 0% | 2,107 | 1,165 | -45% | 0 | 0 | — |
case-10 | pass→pass | 9,819 | 5,153 | -48% | 1 | 1 | 0% | 1,585 | 1,104 | -30% | 0 | 0 | — |
case-11 | fail→fail | 12,624 | 13,636 | +8% | 1 | 1 | 0% | 981 | 1,483 | +51% | 0 | 0 | — |
case-12 | fail→fail | 5,959 | 14,401 | +142% | 1 | 1 | 0% | 714 | 1,262 | +77% | 0 | 0 | — |
case-13 | pass→pass | 20,347 | 8,927 | -56% | 1 | 1 | 0% | 1,999 | 922 | -54% | 0 | 0 | — |
case-14 | fail→fail | 15,663 | 13,738 | -12% | 1 | 1 | 0% | 1,526 | 944 | -38% | 0 | 0 | — |
case-15 | fail→pass | 10,342 | 13,365 | +29% | 1 | 1 | 0% | 658 | 1,608 | +144% | 0 | 0 | — |
case-16 | fail→pass | 15,340 | 4,795 | -69% | 1 | 1 | 0% | 1,721 | 1,183 | -31% | 0 | 0 | — |
case-17 | pass→pass | 16,335 | 7,145 | -56% | 1 | 1 | 0% | 1,659 | 789 | -52% | 0 | 0 | — |
case-18 | fail→fail | 14,794 | 15,925 | +8% | 1 | 1 | 0% | 1,665 | 1,154 | -31% | 0 | 0 | — |
case-19 | pass→pass | 12,566 | 7,730 | -38% | 1 | 1 | 0% | 1,150 | 862 | -25% | 0 | 0 | — |
case-20 | fail→pass | 12,902 | 7,602 | -41% | 1 | 1 | 0% | 934 | 699 | -25% | 0 | 0 | — |
case-21 | fail→fail | 12,443 | 7,614 | -39% | 1 | 1 | 0% | 1,261 | 1,602 | +27% | 0 | 0 | — |
case-22 | fail→fail | 11,687 | 21,421 | +83% | 1 | 1 | 0% | 924 | 3,250 | +252% | 0 | 0 | — |
case-23 | fail→fail | 14,860 | 60,652 | +308% | 1 | 1 | 0% | 1,481 | 8,207 | +454% | 0 | 0 | — |
case-24 | fail→pass | 6,924 | 55,130 | +696% | 1 | 1 | 0% | 876 | 6,007 | +586% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted, and 13 counted toward the lift figure. The other 11 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +17 percentage points is the difference between those two pass rates over the 13 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.