Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute a quick task with GSD guarantees (atomic commits, state tracking) but skip optional agents
.claude/skills/coco-research-gsd-quick/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -77% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -31% | 0% |
<objective> Execute small, ad-hoc tasks with GSD guarantees (atomic commits, STATE.md tracking).
Quick mode is the same system with a shorter path:
.planning/quick/ separate from planned phasesDefault: Skips research, discussion, plan-checker, verifier. Use when you know exactly what to do.
--discuss flag: Lightweight discussion phase before planning. Surfaces assumptions, clarifies gray areas, captures decisions in CONTEXT.md. Use when the task has ambiguity worth resolving upfront.
--full flag: Enables the complete quality pipeline — discussion + research + plan-checking + verification. One flag for everything.
--validate flag: Enables plan-checking (max 2 iterations) and post-execution verification only. Use when you want quality guarantees without discussion or research.
--research flag: Spawns a focused research agent before planning. Investigates implementation approaches, library options, and pitfalls for the task. Use when you're unsure of the best approach.
Granular flags are composable: --discuss --research --validate gives the same result as --full. </objective>
<execution_context> @$HOME/.claude/get-shit-done/workflows/quick.md </execution_context>
<context> $ARGUMENTS
Context files are resolved inside the workflow (init quick) and delegated via <files_to_read> blocks. </context>
<process> Execute the quick workflow from @$HOME/.claude/get-shit-done/workflows/quick.md end-to-end. Preserve all workflow gates (validation, task description, planning, execution, state updates, commits). </process>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 2,942 | 16,249 | +452% | 1 | 1 | 0% | 456 | 718 | +57% | 0 | 0 | — |
case-02 | fail→fail | 14,015 | 11,718 | -16% | 1 | 1 | 0% | 2,120 | 795 | -63% | 0 | 0 | — |
case-03 | fail→fail | 10,487 | 12,636 | +20% | 1 | 1 | 0% | 251 | 830 | +231% | 0 | 0 | — |
case-04 | fail→pass | 10,739 | 2,809 | -74% | 1 | 1 | 0% | 1,318 | 936 | -29% | 0 | 0 | — |
case-05 | fail→pass | 12,874 | 2,874 | -78% | 1 | 1 | 0% | 1,991 | 914 | -54% | 0 | 0 | — |
case-06 | fail→pass | 20,689 | 2,832 | -86% | 1 | 1 | 0% | 3,623 | 825 | -77% | 0 | 0 | — |
case-07 | fail→pass | 11,757 | 3,130 | -73% | 1 | 1 | 0% | 1,889 | 962 | -49% | 0 | 0 | — |
case-08 | pass→pass | 11,601 | 7,469 | -36% | 1 | 1 | 0% | 1,849 | 906 | -51% | 0 | 0 | — |
case-09 | fail→fail | 12,236 | 2,138 | -83% | 1 | 1 | 0% | 1,835 | 704 | -62% | 0 | 0 | — |
case-10 | fail→pass | 17,467 | 10,729 | -39% | 1 | 1 | 0% | 1,795 | 1,237 | -31% | 0 | 0 | — |
case-11 | fail→pass | 20,368 | 15,278 | -25% | 1 | 1 | 0% | 3,330 | 1,664 | -50% | 0 | 0 | — |
case-12 | fail→pass | 8,642 | 7,971 | -8% | 1 | 1 | 0% | 1,450 | 798 | -45% | 0 | 0 | — |
case-13 | pass→pass | 16,342 | 7,874 | -52% | 1 | 1 | 0% | 2,226 | 804 | -64% | 0 | 0 | — |
case-14 | fail→pass | 28,927 | 13,941 | -52% | 1 | 1 | 0% | 4,161 | 1,963 | -53% | 0 | 0 | — |
case-15 | pass→pass | 11,988 | 19,793 | +65% | 1 | 1 | 0% | 2,093 | 3,068 | +47% | 0 | 0 | — |
case-16 | pass→pass | 16,270 | 7,369 | -55% | 1 | 1 | 0% | 1,711 | 823 | -52% | 0 | 0 | — |
case-17 | pass→pass | 13,187 | 5,072 | -62% | 1 | 1 | 0% | 1,267 | 879 | -31% | 0 | 0 | — |
case-18 | pass→pass | 15,944 | 7,866 | -51% | 1 | 1 | 0% | 1,551 | 871 | -44% | 0 | 0 | — |
case-19 | fail→pass | 8,842 | 4,365 | -51% | 1 | 1 | 0% | 1,478 | 943 | -36% | 0 | 0 | — |
case-20 | pass→pass | 18,471 | 9,990 | -46% | 1 | 1 | 0% | 1,824 | 1,303 | -29% | 0 | 0 | — |
case-21 | fail→fail | 18,200 | 15,583 | -14% | 1 | 1 | 0% | 1,995 | 2,194 | +10% | 0 | 0 | — |
case-22 | pass→pass | 14,494 | 10,654 | -26% | 1 | 1 | 0% | 2,146 | 1,337 | -38% | 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, and 19 counted toward the lift figure. The other 3 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 +41 percentage points is the difference between those two pass rates over the 19 comparable cases.
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.