Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Design and facilitate any workshop, working session, or collaborative meeting. Use when asked to plan a workshop, design a facilitated session, run a ideation session, or create a workshop agenda. Produces a complete facilitation guide with session design, activity instructions, timing, and materials.
.claude/skills/mohitagw15856-workshop-facilitation-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 114% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 101% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 86% | 0% |
Produces a complete facilitation guide for any workshop — from a 90-minute problem-solving session to a full-day strategy workshop. Includes step-by-step activity instructions and facilitation moves for when things go off track.
Ask the user for these if not provided:
Date: TBD / as provided] Duration: X hours] Participants: N people, roles] Format: In-person / Remote / Hybrid] Facilitator: Leave for user]
By the end of this session, the group will have:
How we will know it worked: Observable test for success — e.g. "Everyone can name the agreed priorities without looking at their notes"]
Facilitator:
Participants (pre-work):
| Time | Activity | Duration | Format | Output | |---|---|---|---|---| | 00:00] | Welcome and framing | 10 min | Facilitator-led | Shared expectations | | 00:10] | Activity 1] | X min] | Format] | Output] | | 00:X] | Activity 2] | X min] | Format] | Output] | | 00:X] | Break | 15 min | — | — | | 00:X] | Activity 3] | X min] | Format] | Output] | | 00:X] | Decisions and next steps | 20 min | Whole group | Committed actions | | 00:X] | Close | 10 min | Facilitator-led | Energy and commitment |
For each activity:
Purpose: Why this activity at this moment] Time: X minutes] Format: Individual / Pairs / Small groups / Whole group] Materials: Post-its, Miro, printed sheets, etc.]
Instructions to give participants: > "Exact words to say when launching the activity — unambiguous, no jargon]"
Step-by-step:
If the group gets stuck: Specific facilitation move — e.g. "Ask each person to write one idea silently before sharing"] Watch out for: Common failure mode — e.g. "One voice dominating. Use round-robin to surface quieter participants"] Time warning: What to do if running long — e.g. "Skip the prioritisation vote and let facilitator propose the top 3"]
Agree this with the group at the start:
How decisions will be made in this session:
What happens with unresolved disagreements: Parking lot / escalate to sponsor / decide by person] after session]
| Situation | Move | |---|---| | Silence after a question | "Take 2 minutes to write your thoughts before we share" | | One person dominating | "Let's hear from someone we haven't heard from yet" | | Off-topic tangent | "That's important — let me put it in the parking lot. Back to focus]" | | Group stuck, no ideas | "What would competitor / different industry] do here?" | | No consensus, running out of time | "Let's do a quick dot vote to identify the strongest options" | | Energy low after lunch | "Stand up and tell the person next to you your one key takeaway so far" |
End every session with:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 50,683 | 43,525 | -14% | 1 | 1 | 0% | 7,564 | 8,258 | +9% | 0 | 0 | — |
case-02 | fail→fail | 54,667 | 39,928 | -27% | 1 | 1 | 0% | 8,316 | 7,174 | -14% | 0 | 0 | — |
case-03 | pass→pass | 49,781 | 42,177 | -15% | 1 | 1 | 0% | 7,376 | 7,515 | +2% | 0 | 0 | — |
case-04 | fail→pass | 16,640 | 13,663 | -18% | 1 | 1 | 0% | 2,591 | 2,869 | +11% | 0 | 0 | — |
case-05 | fail→fail | 12,288 | 15,565 | +27% | 1 | 1 | 0% | 2,000 | 3,988 | +99% | 0 | 0 | — |
case-06 | fail→pass | 17,016 | 31,924 | +88% | 1 | 1 | 0% | 2,730 | 5,834 | +114% | 0 | 0 | — |
case-07 | fail→fail | 21,283 | 31,664 | +49% | 1 | 1 | 0% | 2,609 | 5,638 | +116% | 0 | 0 | — |
case-08 | fail→pass | 16,093 | 12,344 | -23% | 1 | 1 | 0% | 1,723 | 3,467 | +101% | 0 | 0 | — |
case-09 | pass→pass | 18,203 | 14,016 | -23% | 1 | 1 | 0% | 2,183 | 3,588 | +64% | 0 | 0 | — |
case-18 | pass→pass | 17,334 | 14,726 | -15% | 1 | 1 | 0% | 1,932 | 3,018 | +56% | 0 | 0 | — |
case-10 | pass→pass | 16,010 | 15,204 | -5% | 1 | 1 | 0% | 1,657 | 3,053 | +84% | 0 | 0 | — |
case-11 | pass→pass | 9,099 | 7,963 | -12% | 1 | 1 | 0% | 1,311 | 2,691 | +105% | 0 | 0 | — |
case-12 | pass→pass | 12,578 | 12,028 | -4% | 1 | 1 | 0% | 1,188 | 2,581 | +117% | 0 | 0 | — |
case-13 | fail→pass | 18,014 | 16,498 | -8% | 1 | 1 | 0% | 1,928 | 3,250 | +69% | 0 | 0 | — |
case-23 | pass→pass | 21,393 | 23,598 | +10% | 1 | 1 | 0% | 2,687 | 5,393 | +101% | 0 | 0 | — |
case-14 | pass→pass | 20,783 | 19,287 | -7% | 1 | 1 | 0% | 2,526 | 3,798 | +50% | 0 | 0 | — |
case-15 | pass→pass | 19,952 | 20,924 | +5% | 1 | 1 | 0% | 2,071 | 3,849 | +86% | 0 | 0 | — |
case-16 | pass→pass | 18,269 | 23,424 | +28% | 1 | 1 | 0% | 2,149 | 4,342 | +102% | 0 | 0 | — |
case-17 | fail→pass | 15,963 | 11,750 | -26% | 1 | 1 | 0% | 1,798 | 3,336 | +86% | 0 | 0 | — |
case-19 | pass→pass | 11,781 | 11,983 | +2% | 1 | 1 | 0% | 1,841 | 2,654 | +44% | 0 | 0 | — |
case-20 | pass→pass | 15,386 | 25,296 | +64% | 1 | 1 | 0% | 2,526 | 4,766 | +89% | 0 | 0 | — |
case-21 | pass→pass | 15,919 | 11,054 | -31% | 1 | 1 | 0% | 1,432 | 3,447 | +141% | 0 | 0 | — |
case-22 | pass→pass | 13,709 | 9,718 | -29% | 1 | 1 | 0% | 1,448 | 3,196 | +121% | 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. 23 cases were attempted. The headline lift of +22 percentage points is the difference between those two pass rates over the 23 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.