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 applies the slide source hierarchy (pptx > pdf > video_extracted > none), treats video_url as the processability gate, handles fallback chains, and sets correct status per the error handling matrix.",
"type": "weighted_checklist",
"checklist": [
{
"name": "PPTX highest priority",
"description": "When both PPTX and PDF are available (Microservices Myths), slide_source reflects that both are available — not just PDF",
"max_score": 10
},
{
"name": "Video-only talks are processable",
"description": "CI Pipeline Masterclass (video_url but no slides_url, no pptx_path) is NOT skipped — it gets slide_source indicating video extraction and can reach a processed status",
"max_score": 12
},
{
"name": "No-video talks not processable",
"description": "Testing in Production (has slides_url but no video_url) is marked as not processable / skipped — video_url is required to enter the processing pipeline",
"max_score": 10
},
{
"name": "Multiple slide_source values used",
"description": "The output database uses at least 3 distinct slide_source values across the 8 talks — reflecting the different acquisition paths",
"max_score": 10
},
{
"name": "Video extraction output named by YouTube ID",
"description": "For video-extracted talks, the output path or filename references the YouTube video ID (e.g., 'ghi789.pdf'), not a generic name",
"max_score": 8
},
{
"name": "Slides fail + video available = fallback to video",
"description": "DevOps Culture Reset (slides download failed but has video_url) attempts video extraction as fallback — the processing log shows the fallback chain",
"max_score": 10
},
{
"name": "Partial processing when one source fails",
"description": "Talks where transcript or slides (but not both) fail get a status distinct from full success — indicating partial analysis is possible",
"max_score": 10
},
{
"name": "All fail = distinct failure status",
"description": "GitOps Journey (both transcript and slides failed) gets a status indicating download failure, distinct from the missing-sources skip",
"max_score": 10
},
{
"name": "Processing log explains decisions",
"description": "The log explains the source selection rationale for at least 5 talks — not just listing outcomes but explaining why each source was chosen",
"max_score": 20
}
]
}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