Post-delivery evaluation of a keynote. Did the provocation land? Did the audience shift? What to iterate for next time. Use for "how did my talk go", "keynote eval", "evaluate my speech".
From uxinatornpx claudepluginhub jabberlockie/the-human-stack-plugins-public --plugin uxinatorThis skill uses the workspace's default tool permissions.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Details PluginEval's skill quality evaluation: 3 layers (static, LLM judge), 10 dimensions, rubrics, formulas, anti-patterns, badges. Use to interpret scores, improve triggering, calibrate thresholds.
Post-delivery retrospective for keynotes and conference talks. Assesses whether the provocation landed, the emotional arc worked, and the audience shifted. Feeds directly into speech-iterate for versioning the talk.
DID THE AUDIENCE SHIFT?
───────────────────────
BEFORE STATE: [What the audience believed/felt before the talk]
INTENDED AFTER STATE: [What they should believe/feel after]
ACTUAL AFTER STATE: [Evidence of what actually happened]
SHIFT ACHIEVED: [YES | PARTIALLY | NO]
EVIDENCE:
- [Quote, behavior, feedback that shows shift]
- [Quote, behavior, feedback that shows shift]
- [Or: silence / polite applause / generic "that was great" = NO shift evidence]
THE ONE LINE
────────────
DESIGNED LINE: [What was supposed to be repeated]
DID ANYONE REPEAT IT? [YES — where? | NO]
WHAT THEY ACTUALLY REPEATED: [Sometimes a different line lands harder]
IF DIFFERENT: Consider making THAT line the designed one in v2.
PEAK-END EVALUATION
───────────────────
DESIGNED PEAK: [What was planned as the peak moment]
ACTUAL PEAK: [What the audience responded to most intensely]
MATCH: [SAME | DIFFERENT — note what surprised]
DESIGNED CLOSE: [What was planned]
ACTUAL CLOSE: [What Tim actually said/did]
CLOSE QUALITY: [STRONG | ADEQUATE | TRAILED OFF | DIFFERENT FROM PLAN]
PEAK-END PREDICTION: [Based on actual peak + actual end, what will
the audience REMEMBER about this talk?]
MODE DISCIPLINE
───────────────
PLANNED MODE: KEYNOTE (provocation)
ACTUAL MODE: [What Tim actually did — may vary by section]
DRIFT MOMENTS:
- [Timestamp/section]: [What happened — e.g., "Stopped to explain the
UX² framework for 4 minutes. Audience energy dropped visibly."]
EXPLAIN CONTAMINATION: [X]% of delivery was education, not provocation
AUDIENCE RESPONSE DATA
──────────────────────
DURING TALK:
- Energy shifts: [When did they lean in? When did they check phones?]
- Laughter: [Where? Was it earned or polite?]
- Note-taking: [Where? This indicates "useful" not "moved"]
- Visible reactions: [Nods, head shakes, whispers to neighbor]
AFTER TALK:
- Approach rate: [How many people came up to Tim after?]
- Questions asked: [What did they ask? "How do I..." = training response.
"I never thought of it that way" = provocation response.]
- Social mentions: [Any posts, tags, quotes shared?]
- Feedback forms: [Scores + verbatim comments]
- Invitations: [Did anyone invite Tim to speak again? Where?]
- Connections: [LinkedIn requests, card exchanges, DMs]
SIGNAL INTERPRETATION:
[What do these signals tell us about whether the talk provoked or taught?]
TIM'S FELT EXPERIENCE
─────────────────────
Energy level: [How did Tim feel during delivery? Alive? Flat? Rushed?]
Story delivery: [Did the anchor story land the way he wanted?]
Close execution: [Did Tim deliver the close as designed or improvise?]
Drift awareness: [Did Tim notice himself drifting? Where?]
Audience connection: [Did Tim feel connected to the room or performing at it?]
Surprise moments: [Anything unexpected that worked? That didn't?]
KEYNOTE EVALUATION
══════════════════
PROVOCATION: [LANDED | PARTIALLY | MISSED]
EMOTIONAL ARC: [COMPELLING | ADEQUATE | FLAT]
CLOSE: [STRONG | ADEQUATE | WEAK]
MODE DISCIPLINE: [CLEAN | MINOR DRIFT | MAJOR DRIFT]
AUDIENCE SHIFT: [EVIDENT | UNCLEAR | NO EVIDENCE]
OVERALL: [X/50 across 5 dimensions, 10 each]
WHAT WORKED (keep for v2):
1. [Specific]
2. [Specific]
WHAT DIDN'T (fix for v2):
1. [Specific]
2. [Specific]
VERSION NOTE: [What should change in the next delivery of this talk?
Feed this into speech-iterate.]