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 4-tier pattern strategy from the speaker's pattern profile, with antipatterns tagged as [RECURRING] or [CONTEXTUAL] based on the speaker's history.",
"type": "weighted_checklist",
"checklist": [
{
"name": "4-tier structure present",
"description": "The report organizes pattern recommendations into exactly four labeled tiers: Signature (or YOUR TOOLKIT), Contextual (or WORTH CONSIDERING), New to You (or NEW TO YOU), and Shake It Up (or WILD CARD) — four distinct sections, not three or five",
"max_score": 20
},
{
"name": "Signature tier from profile",
"description": "The Signature tier contains patterns from the speaker's signature_patterns (narrative-arc, foreshadowing, brain-breaks, bookends) — not invented patterns",
"max_score": 12
},
{
"name": "Contextual tier from profile",
"description": "The Contextual tier contains at least one pattern from the speaker's contextual_patterns (expansion-joints, a-la-carte-content, talklet)",
"max_score": 10
},
{
"name": "Shake It Up has 1-2 items",
"description": "The Shake It Up / Wild Card tier contains exactly 1 or 2 pattern suggestions — not 0 and not 3 or more",
"max_score": 12
},
{
"name": "[RECURRING] tag used",
"description": "At least one antipattern detection is labeled [RECURRING] (matching shortchanged or bullet-riddled-corpse from the speaker's antipattern_frequency history)",
"max_score": 15
},
{
"name": "[CONTEXTUAL] tag used",
"description": "At least one antipattern detection is labeled [CONTEXTUAL] — a risk detected in the current outline that is NOT in the speaker's historical antipattern_frequency (e.g., meme accretion in the opening sequence)",
"max_score": 15
},
{
"name": "New to You from never_used",
"description": "The New to You tier suggests at least one pattern from the speaker's never_used_patterns list — not patterns already in signature or contextual tiers",
"max_score": 8
},
{
"name": "Actionable recommendations",
"description": "At least 2 pattern recommendations include specific advice for how to apply them to THIS talk — not generic definitions of what the pattern is",
"max_score": 8
}
]
}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