Four-skill presentation system: ingest talks into a rhetoric vault, run interactive clarification, generate a speaker profile, then create new presentations that match your documented patterns. Includes an 88-entry Presentation Patterns taxonomy for scoring, brainstorming, and go-live preparation.
96
93%
Does it follow best practices?
Impact
97%
1.21xAverage score across 30 eval scenarios
Advisory
Suggest reviewing before use
{
"context": "Tests whether the agent produces a talk analysis following the skill's 14-dimension framework, structured data schema, and verbatim examples extraction requirements.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Opening pattern analyzed",
"description": "Analysis identifies and categorizes the opening pattern (e.g., as provocative/bold claim, audience interaction, or similar)",
"max_score": 6
},
{
"name": "Narrative structure analyzed",
"description": "Analysis describes the narrative arc type (e.g., problem-diagnosis-solution, chronological, or listicle)",
"max_score": 6
},
{
"name": "Humor documented",
"description": "Analysis identifies specific humor instances, categorizing them by type (self-deprecating, meme-based, etc.)",
"max_score": 6
},
{
"name": "Audience interaction noted",
"description": "Analysis identifies audience engagement techniques (the 'raise your hand' moments in the transcript)",
"max_score": 6
},
{
"name": "Areas for improvement",
"description": "Analysis includes an honest critical assessment section identifying weaknesses (e.g., rushed ending indicated by 'five minute warning')",
"max_score": 8
},
{
"name": "Structured data fields",
"description": "The JSON output includes structured_data with at least: slide_count, opening_type, closing_type, narrative_arc_type, and audience_interaction_count",
"max_score": 10
},
{
"name": "Opening type categorized correctly",
"description": "opening_type correctly captures that this talk opens with a bold/provocative claim combined with audience polling — the value should be semantically accurate, not just any placeholder",
"max_score": 8
},
{
"name": "Closing type categorized correctly",
"description": "closing_type correctly captures that this talk closes with a numbered summary plus call-to-action and social handles — the value should reflect the actual closing structure",
"max_score": 8
},
{
"name": "Verbatim signature phrases",
"description": "Output includes a verbatim_examples section with signature_phrases containing actual recurring expressions from the transcript (e.g., 'right', 'okay so')",
"max_score": 8
},
{
"name": "Verbatim opening lines",
"description": "Output includes opening_lines with the actual first few sentences from the transcript, quoted verbatim",
"max_score": 8
},
{
"name": "Verbatim closing lines",
"description": "Output includes closing_lines with the actual last few sentences from the transcript, quoted verbatim",
"max_score": 8
},
{
"name": "Transition phrases extracted",
"description": "Output includes actual transition phrases from the transcript in verbatim_examples.transitions (e.g., 'next thing you know', 'jokes aside', 'okay so')",
"max_score": 8
},
{
"name": "JSON schema completeness",
"description": "The top-level JSON includes all required fields: filename, rhetoric_notes, areas_for_improvement, structured_data, verbatim_examples, new_patterns, summary_updates",
"max_score": 10
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17
scenario-18
scenario-19
scenario-20
scenario-21
scenario-22
scenario-23
scenario-24
scenario-25
scenario-26
scenario-27
scenario-28
scenario-29
scenario-30
rules
skills
presentation-creator
references
patterns
build
deliver
prepare
scripts
vault-clarification
vault-ingress
vault-profile