mikeparcewski/wicked-garden / skills/jam
wicked-garden-jam
VerifiedOrchestrates AI-powered brainstorming sessions with dynamic focus groups. quick sessions are ephemeral (no storage). brainstorm and council sessions are tracked as native tasks (process) and stored via wicked-garden-mem (outcome). Use when: "brainstorm this", "explore ideas", "get different perspectives", "focus group", "what do you think about", "pros and cons", "quick check", "jam quick <idea>", "jam brainstorm <topic>", "jam council <topic> with options A/B/C", "multi-model evaluation", "council verdict", "jam revisit <decision>", "how did that decision work out".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.5 KB |
| refs/council-verdict.md | markdown | 1.9 KB |
| refs/facilitation-patterns.md | markdown | 5.4 KB |
| refs/quick.md | markdown | 2.1 KB |
| refs/revisit.md | markdown | 2.2 KB |
| refs/synthesis-patterns.md | markdown | 4.4 KB |
refs/synthesis-patterns.md
View on GitHub---
phase_relevance: ["clarify", "design"]
archetype_relevance: ["*"]
---
# Synthesis Patterns
How to transform raw persona discussions into actionable outcomes.
## Synthesis Structure
Every synthesis follows this skeleton:
```markdown
## Brainstorm: {Topic}
### Key Insights
1. **{Insight}** — {confidence: HIGH/MEDIUM/LOW}
{1-2 sentences explaining the non-obvious finding}
2. ...
3. ...
### Action Items
1. {Specific, actionable next step} — {who/when if known}
2. ...
### Open Questions
- {What remains unresolved — honest about gaps}
### Tensions
- **{Position A}** vs **{Position B}** — {why this tension matters}
```
## Quality Checklist
A good synthesis has:
- [ ] 3-5 insights (not 10 — information overload kills action)
- [ ] Confidence levels on every insight (HIGH/MEDIUM/LOW)
- [ ] At least 1 non-obvious finding (something no single persona said alone)
- [ ] Prioritized action items (not just a flat list)
- [ ] Honest open questions (admitting what's unknown)
- [ ] Named tensions (disagreements that matter, not smoothed over)
A bad synthesis:
- Just lists what each persona said (that's a transcript, not synthesis)
- Has 8+ insights with no prioritization
- Everything is "MEDIUM" confidence (lazy scoring)
- No action items (interesting but not useful)
- No tensions acknowledged (false consensus)
## Synthesis Techniques
### The Non-Obvious Connection
Look for ideas that emerge from the intersection of two personas' viewpoints — something neither would have said alone.
**Example**: The engineer says "we need a cache layer" and the user researcher says "users hate stale data." The synthesis insight is: "Cache with aggressive invalidation — latency matters but correctness matters more." Neither persona stated this directly.
### The Surprising Agreement
When personas who should disagree actually agree on something, that's a high-confidence signal.
**Example**: The skeptic and the visionary both say "this needs to be simple." That's a stronger signal than either alone.
### The Productive Tension
Some disagreements shouldn't be resolved — they should be named as trade-offs for the decision-maker.
**Example**: "Speed of delivery vs. test coverage. The team lead wants to ship Friday; QE wants integration tests. This is a risk decision, not a right answer."
### The Missing Voice
If no persona raised a concern that seems obvious, note it. The panel may have been wrong for this problem.
**Example**: "No persona discussed data privacy implications. This gap should be addressed before implementation."
## Decision Record Format
When a brainstorm produces a clear decision, store it via the wicked-garden-mem skill (store action):
```
Decision: {what was decided}
Chosen: {the selected option}
Rationale: {why — rooted in evidence from the session}
Alternatives considered: {what was rejected and why}
Confidence: {HIGH/MEDIUM/LOW}
Evidence used: {what data/experience informed the decision}
Personas: {who participated}
Tags: {topic tags for future recall}
```
This format enables future recall: "What did we decide about X?" returns the full decision context.
## Examples of Good Synthesis
### Architecture Decision
> **Key Insight 1**: Event sourcing is the right pattern for audit requirements, but the team has no experience with it — HIGH confidence on the pattern choice, LOW confidence on execution timeline.
>
> **Action**: Spike on event sourcing with a bounded context (2 days) before committing to full adoption.
>
> **Tension**: Architect wants event sourcing everywhere; pragmatist wants it only for the audit domain. Resolve after the spike.
### Product Scope
> **Key Insight 1**: The top 3 requested features serve power users, but 80% of new signups churn before reaching power-user status — HIGH confidence.
>
> **Action**: Prioritize onboarding improvements over feature additions for Q2.
>
> **Open Question**: What does the conversion funnel look like between signup and first value moment? No persona had this data.
### Process Change
> **Key Insight 1**: The team unanimously agrees code review is too slow, but disagrees on the fix. The bottleneck is not the review process — it's that PRs are too large — HIGH confidence.
>
> **Action**: Adopt a "300 lines max" PR policy for 2 sprints as an experiment. Measure review turnaround.
>
> **Tension**: ICs want autonomy on PR size; lead wants consistency. The experiment resolves this empirically.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |