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 surfaces a missing shortener configuration instead of silently degrading to a raw URL. Missing config is not the same as intentional opt-out. Also tests that the shownotes URL slug comes from the Presentation Spec, not agent-invented.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Agent surfaces missing shortener config",
"description": "The agent explicitly identifies that the shortener key is missing from the speaker profile's qr_code config and communicates this to the user. Must not silently proceed with a raw URL",
"max_score": 20
},
{
"name": "Agent asks user to choose a shortener",
"description": "The agent asks the user to choose between bitly, rebrandly, or explicitly 'none' — presenting it as a decision, not a silent default. The agent does not assume 'none' on behalf of the user",
"max_score": 15
},
{
"name": "Agent does NOT generate QR with raw URL without asking",
"description": "The agent does not produce a QR code encoding the raw shownotes URL without first getting explicit user confirmation that no shortener is desired. Silent raw-URL generation = zero points",
"max_score": 15
},
{
"name": "Shownotes URL uses slug from Presentation Spec",
"description": "The shownotes URL is constructed as https://jbaru.ch/2026-04-16-devnexus-robocoders-judgment-day, using the exact slug from presentation-spec.md. The agent does not invent, abbreviate, or rephrase the slug",
"max_score": 15
},
{
"name": "Talk slug matches spec exactly",
"description": "Any reference to the talk slug uses exactly '2026-04-16-devnexus-robocoders-judgment-day' as written in the Presentation Spec, not a shortened or modified version",
"max_score": 10
},
{
"name": "Agent references generate-qr.py as the tool",
"description": "The agent's plan references using generate-qr.py (the skill script), not direct use of the qrcode library or any ad-hoc QR generation approach",
"max_score": 10
},
{
"name": "Agent distinguishes missing config from opt-out",
"description": "The agent's response shows it understands the distinction: 'shortener key missing' means NOT CONFIGURED (needs a decision), not the same as 'shortener: none' (explicit opt-out)",
"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