Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cleanly stop an active engineering layer flow (dev-autopilot, dev-deep-interview, dev-plan) and report what was completed. Use when the user says "cancel", "stop", "abort", "cancelomc", or wants to exit a multi-phase dev workflow.
.claude/skills/evolution-foundation-dev-cancel/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 81% | 0% |
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer — simplified, runtime-free version.
Stop the active flow cleanly and give the user a structured summary of:
Look at recent context for signals:
dev-autopilot was invoked → multi-phase pipeline activedev-deep-interview was invoked → interview loop activedev-plan was invoked → planning session active@compass-planner, @bolt-executor, etc. agent was runningOutput this structure:
markdown## Engineering Layer Cancel Report **Flow stopped:** {dev-autopilot | dev-deep-interview | dev-plan | other} **Stopped at phase:** {phase name} **Reason:** {user request / error / explicit abort} ### Completed - [phase 1: spec written → workspace/projects/specs/...] - [phase 2: plan approved → workspace/projects/plans/...] ### In Progress (at cancel time) - [phase X: implementing Y in file Z] ### Not Started - [remaining phases] ### Artifacts Preserved - Spec: `workspace/projects/specs/[C]autopilot-spec-{name}.md` - Plan: `workspace/projects/plans/[C]autopilot-plan-{name}.md` - Code changes: see `git status` ### Recommended Next Step - "Resume manually": pick up from {phase} using {agent} directly - "Re-run from scratch": delete artifacts and re-invoke `dev-autopilot` - "Investigate first": review the artifacts before deciding ### Open Questions [any unresolved decisions the flow was waiting on]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→pass | 8,031 | 5,230 | -35% | 1 | 1 | 0% | 1,209 | 1,663 | +38% | 0 | 0 | — |
case-06 | fail→pass | 8,373 | 5,343 | -36% | 1 | 1 | 0% | 1,352 | 1,648 | +22% | 0 | 0 | — |
case-07 | fail→fail | 14,116 | 7,133 | -49% | 1 | 1 | 0% | 1,277 | 2,140 | +68% | 0 | 0 | — |
case-01 | fail→pass | 7,871 | 5,358 | -32% | 1 | 1 | 0% | 1,330 | 1,754 | +32% | 0 | 0 | — |
case-02 | fail→fail | 6,926 | 3,412 | -51% | 1 | 1 | 0% | 1,129 | 1,385 | +23% | 0 | 0 | — |
case-03 | fail→pass | 5,717 | 3,744 | -35% | 1 | 1 | 0% | 924 | 1,425 | +54% | 0 | 0 | — |
case-04 | fail→pass | 6,810 | 6,910 | +1% | 1 | 1 | 0% | 1,118 | 2,028 | +81% | 0 | 0 | — |
case-16 | pass→pass | 11,822 | 5,967 | -50% | 1 | 1 | 0% | 1,880 | 1,888 | +0% | 0 | 0 | — |
case-08 | fail→pass | 1,581 | 4,940 | +212% | 1 | 1 | 0% | 177 | 1,621 | +816% | 0 | 0 | — |
case-09 | fail→pass | 3,711 | 5,780 | +56% | 1 | 1 | 0% | 568 | 1,800 | +217% | 0 | 0 | — |
case-10 | pass→pass | 7,500 | 3,021 | -60% | 1 | 1 | 0% | 1,165 | 1,280 | +10% | 0 | 0 | — |
case-11 | pass→fail | 8,371 | 7,020 | -16% | 1 | 1 | 0% | 1,277 | 1,908 | +49% | 0 | 0 | — |
case-12 | fail→pass | 7,484 | 6,689 | -11% | 1 | 1 | 0% | 1,126 | 2,024 | +80% | 0 | 0 | — |
case-13 | fail→fail | 4,779 | 7,812 | +63% | 1 | 1 | 0% | 721 | 2,059 | +186% | 0 | 0 | — |
case-14 | pass→pass | 6,016 | 8,251 | +37% | 1 | 1 | 0% | 1,033 | 2,278 | +121% | 0 | 0 | — |
case-15 | pass→pass | 2,824 | 2,685 | -5% | 1 | 1 | 0% | 376 | 1,183 | +215% | 0 | 0 | — |
case-17 | pass→pass | 11,017 | 4,961 | -55% | 1 | 1 | 0% | 1,712 | 1,639 | -4% | 0 | 0 | — |
case-18 | pass→pass | 12,733 | 4,107 | -68% | 1 | 1 | 0% | 1,946 | 1,453 | -25% | 0 | 0 | — |
case-19 | pass→pass | 9,656 | 2,467 | -74% | 1 | 1 | 0% | 1,539 | 1,144 | -26% | 0 | 0 | — |
case-20 | pass→pass | 3,942 | 4,014 | +2% | 1 | 1 | 0% | 518 | 1,437 | +177% | 0 | 0 | — |
case-21 | pass→pass | 3,219 | 2,669 | -17% | 1 | 1 | 0% | 457 | 1,197 | +162% | 0 | 0 | — |
case-22 | pass→pass | 3,881 | 5,668 | +46% | 1 | 1 | 0% | 514 | 1,754 | +241% | 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. 22 cases were attempted. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 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.