Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Turn a plan or set of milestones into a timeline / Gantt chart. Use when asked to build a roadmap, schedule phases, show a project timeline, or visualize what happens when. Produces a ready-to-render Mermaid Gantt chart (renders live, exportable as PNG/SVG) — and, because it has real dates, the result also exports to a calendar (.ics) — plus notes on the critical path and risks.
.claude/skills/mohitagw15856-gantt-roadmap/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 59% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 75% | 0% |
A list of tasks doesn't show what runs in parallel, what blocks what, or where the crunch is. A Gantt chart does. This skill turns a plan into a Mermaid Gantt chart with phases (sections), dated tasks, dependencies, and milestones — a real schedule, not a wish list. Because the output carries real dates, the playground can also export it straight to a calendar (.ics).
Ask for these only if they aren't already provided:
If exact dates aren't given, anchor to a start date and lay tasks out by stated duration/order; flag the dates as planning estimates.
One line on the time span and goal.
mermaidgantt title [Project] roadmap dateFormat YYYY-MM-DD axisFormat %b %d section Discovery Research :done, r1, 2026-07-01, 10d Spec sign-off :milestone, m1, 2026-07-15, 0d section Build Core build :active, b1, after m1, 20d Integrations : b2, after b1, 10d section Launch Beta :milestone, m2, 2026-08-25, 0d GA :milestone, m3, 2026-09-10, 0d
Critical path — the chain of dependent tasks that sets the end date.
Risks / buffers — where the schedule is tight, what could slip, where buffer exists.
Assumptions — any dates you estimated rather than were given.
gantt, then title, dateFormat YYYY-MM-DD, optional axisFormat.section Name. Task line: Label : [status,] id, start, duration (e.g. :active, b1, 2026-07-01, 20d).after <id> as the start. Milestones use the milestone tag with 0d.YYYY-MM-DD) so the calendar (.ics) export works.after so the schedule reflects what blocks whatYYYY-MM-DDProject scheduling (Gantt charts, critical path, milestones, dependencies), expressed as renderable Mermaid.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 14,721 | 11,385 | -23% | 1 | 1 | 0% | 3,119 | 3,187 | +2% | 0 | 0 | — |
case-02 | fail→fail | 15,830 | 11,457 | -28% | 1 | 1 | 0% | 3,458 | 3,305 | -4% | 0 | 0 | — |
case-03 | pass→pass | 8,074 | 8,831 | +9% | 1 | 1 | 0% | 1,773 | 2,726 | +54% | 0 | 0 | — |
case-04 | fail→pass | 6,880 | 7,056 | +3% | 1 | 1 | 0% | 1,599 | 2,468 | +54% | 0 | 0 | — |
case-05 | pass→pass | 4,312 | 7,261 | +68% | 1 | 1 | 0% | 992 | 2,312 | +133% | 0 | 0 | — |
case-06 | pass→pass | 8,970 | 7,831 | -13% | 1 | 1 | 0% | 2,030 | 2,526 | +24% | 0 | 0 | — |
case-07 | fail→fail | 13,045 | 10,791 | -17% | 1 | 1 | 0% | 2,573 | 3,097 | +20% | 0 | 0 | — |
case-08 | pass→pass | 8,413 | 6,220 | -26% | 1 | 1 | 0% | 1,962 | 2,323 | +18% | 0 | 0 | — |
case-09 | pass→pass | 14,718 | 10,324 | -30% | 1 | 1 | 0% | 2,684 | 2,913 | +9% | 0 | 0 | — |
case-10 | pass→fail | 13,494 | 16,968 | +26% | 1 | 1 | 0% | 2,470 | 4,320 | +75% | 0 | 0 | — |
case-11 | pass→fail | 7,907 | 9,411 | +19% | 1 | 1 | 0% | 1,642 | 2,841 | +73% | 0 | 0 | — |
case-12 | pass→fail | 10,811 | 13,958 | +29% | 1 | 1 | 0% | 1,824 | 3,451 | +89% | 0 | 0 | — |
case-13 | pass→pass | 8,229 | 10,778 | +31% | 1 | 1 | 0% | 1,764 | 3,141 | +78% | 0 | 0 | — |
case-14 | fail→pass | 11,816 | 8,739 | -26% | 1 | 1 | 0% | 2,594 | 2,702 | +4% | 0 | 0 | — |
case-15 | pass→pass | 4,440 | 6,261 | +41% | 1 | 1 | 0% | 1,074 | 2,181 | +103% | 0 | 0 | — |
case-16 | pass→pass | 4,244 | 6,467 | +52% | 1 | 1 | 0% | 959 | 2,277 | +137% | 0 | 0 | — |
case-17 | pass→pass | 6,290 | 7,503 | +19% | 1 | 1 | 0% | 1,433 | 2,419 | +69% | 0 | 0 | — |
case-18 | fail→pass | 9,408 | 8,007 | -15% | 1 | 1 | 0% | 2,198 | 2,463 | +12% | 0 | 0 | — |
case-19 | fail→pass | 8,421 | 8,284 | -2% | 1 | 1 | 0% | 1,774 | 2,829 | +59% | 0 | 0 | — |
case-20 | pass→pass | 13,164 | 10,448 | -21% | 1 | 1 | 0% | 2,888 | 3,133 | +8% | 0 | 0 | — |
case-21 | fail→fail | 13,194 | 10,920 | -17% | 1 | 1 | 0% | 2,587 | 2,864 | +11% | 0 | 0 | — |
case-22 | pass→pass | 8,297 | 7,029 | -15% | 1 | 1 | 0% | 1,664 | 2,328 | +40% | 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 +5 percentage points is the difference between those two pass rates over the 22 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.