Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Run all remaining phases autonomously — discuss→plan→execute per phase
.claude/skills/davepoon-gsd-autonomous/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -64% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -60% | 0% |
<objective> Execute all remaining milestone phases autonomously. For each phase: discuss → plan → execute. Pauses only for user decisions (grey area acceptance, blockers, validation requests).
Uses ROADMAP.md phase discovery and Skill() flat invocations for each phase command. After all phases complete: milestone audit → complete → cleanup.
Creates/Updates:
.planning/STATE.md — updated after each phase.planning/ROADMAP.md — progress updated after each phaseAfter: Milestone is complete and cleaned up. </objective>
<execution_context> @${CLAUDE_PLUGIN_ROOT}/workflows/autonomous.md @${CLAUDE_PLUGIN_ROOT}/references/ui-brand.md </execution_context>
<context> Optional flags:
--from N — start from phase N instead of the first incomplete phase.--to N — stop after phase N completes (halt instead of advancing to next phase).--only N — execute only phase N (single-phase mode).--interactive — run discuss inline with questions (not auto-answered), then dispatch plan→execute as background agents. Keeps the main context lean while preserving user input on decisions.Project context, phase list, and state are resolved inside the workflow using init commands (gsd-sdk query init.milestone-op, gsd-sdk query roadmap.analyze). No upfront context loading needed. </context>
<process> Execute the autonomous workflow from @${CLAUDE_PLUGIN_ROOT}/workflows/autonomous.md end-to-end. Preserve all workflow gates (phase discovery, per-phase execution, blocker handling, progress display). </process>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-14 | fail→pass | 8,594 | 1,785 | -79% | 1 | 1 | 0% | 1,424 | 650 | -54% | 0 | 0 | — |
case-19 | pass→pass | 8,798 | 2,627 | -70% | 1 | 1 | 0% | 1,482 | 777 | -48% | 0 | 0 | — |
case-01 | fail→fail | 6,647 | 5,667 | -15% | 1 | 1 | 0% | 1,179 | 664 | -44% | 0 | 0 | — |
case-02 | pass→fail | 8,341 | 4,874 | -42% | 1 | 1 | 0% | 1,535 | 704 | -54% | 0 | 0 | — |
case-03 | fail→fail | 4,378 | 6,975 | +59% | 1 | 1 | 0% | 727 | 937 | +29% | 0 | 0 | — |
case-04 | pass→fail | 9,905 | 6,597 | -33% | 1 | 1 | 0% | 1,571 | 785 | -50% | 0 | 0 | — |
case-05 | pass→fail | 11,156 | 5,302 | -52% | 1 | 1 | 0% | 1,917 | 1,228 | -36% | 0 | 0 | — |
case-06 | fail→fail | 3,458 | 3,290 | -5% | 1 | 1 | 0% | 590 | 921 | +56% | 0 | 0 | — |
case-07 | fail→fail | 3,427 | 5,266 | +54% | 1 | 1 | 0% | 501 | 733 | +46% | 0 | 0 | — |
case-08 | fail→fail | 4,810 | 3,483 | -28% | 1 | 1 | 0% | 754 | 587 | -22% | 0 | 0 | — |
case-09 | fail→fail | 4,462 | 5,406 | +21% | 1 | 1 | 0% | 630 | 769 | +22% | 0 | 0 | — |
case-10 | fail→pass | 12,658 | 3,118 | -75% | 1 | 1 | 0% | 2,286 | 833 | -64% | 0 | 0 | — |
case-11 | fail→pass | 6,476 | 1,575 | -76% | 1 | 1 | 0% | 1,046 | 662 | -37% | 0 | 0 | — |
case-12 | fail→pass | 11,557 | 4,791 | -59% | 1 | 1 | 0% | 2,369 | 1,209 | -49% | 0 | 0 | — |
case-13 | pass→pass | 9,106 | 4,707 | -48% | 1 | 1 | 0% | 1,681 | 1,190 | -29% | 0 | 0 | — |
case-15 | fail→pass | 10,107 | 1,778 | -82% | 1 | 1 | 0% | 1,667 | 673 | -60% | 0 | 0 | — |
case-16 | fail→pass | 12,372 | 2,064 | -83% | 1 | 1 | 0% | 2,109 | 614 | -71% | 0 | 0 | — |
case-17 | pass→pass | 11,449 | 3,199 | -72% | 1 | 1 | 0% | 1,864 | 912 | -51% | 0 | 0 | — |
case-18 | fail→pass | 9,734 | 4,073 | -58% | 1 | 1 | 0% | 1,662 | 966 | -42% | 0 | 0 | — |
case-20 | fail→pass | 14,242 | 7,305 | -49% | 1 | 1 | 0% | 2,380 | 1,607 | -32% | 0 | 0 | — |
case-21 | fail→pass | 8,965 | 6,672 | -26% | 1 | 1 | 0% | 1,515 | 1,541 | +2% | 0 | 0 | — |
case-22 | fail→pass | 10,066 | 1,659 | -84% | 1 | 1 | 0% | 1,719 | 604 | -65% | 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 15 counted toward the lift figure. The other 7 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 +32 percentage points is the difference between those two pass rates over the 15 comparable cases. 3 cases got worse with the skill loaded, and they are 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.