talk-scheire-artificial-intelligence

Use when the user asks about Lieven Scheire's talk "Artificial Intelligence" (a Belgian physicist/comedian's keynote on AI for a developer audience) — including questions about his one-sentence definition of AI as "a new kind of software good at pattern recognition", the history of AI from the 1956 Dartmouth workshop, how neural networks mimic the brain, training-data bias (the "snow in the background" wolves-vs-huskies example, the dermatology ruler example), the black-box nature of neural nets, hobbyist AI tools (Teachable Machine, Custom Vision, HeyGen, PhotoMath, Merlin), Ben Hamm's cat flap, his skeptical stance on LLMs as "language imitation" vs AGI, safe excerpts from the talk, or applying his framing to current AI work.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./Plugins/aidevcon/skills/talk-scheire-artificial-intelligence/SKILL.md

Quality

Content

63%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-organized, actionable reference/Q&A skill with a clear retrieval workflow and concrete attribution guardrails, undermined by a critical bundle defect: its entire workflow depends on outline.md, transcript.md, and quote.md, none of which ship with the skill.

Suggestions

Ship the referenced bundle files (outline.md, transcript.md, quote.md) under references/ so the lookup procedure can actually be executed; the SKILL.md is currently a pointer to absent resources.

Merge the duplicated workflow steps in "Grounding rules" and "General lookup procedure" into a single canonical procedure to remove the conciseness redundancy.

Add a quick pointer in the body to where outline.md/transcript.md/quote.md live and their roles, so a reader can confirm the disclosure layer exists before relying on it.

Dimension	Reasoning	Score
Conciseness	The body is task-specific and avoids explaining concepts Claude already knows, but the "Grounding rules" (items 1-4) and the "General lookup procedure" (steps 1-5) restate the same read-outline → read-transcript → cite → say-if-absent workflow twice, which could be tightened.	3 / 5
Actionability	Concrete, specific guidance throughout — exact files to read, exact fallback phrasing ("the talk doesn't address this"), explicit attribution phrasings ("Scheire said...", "the host said..."), and per-scenario procedures covering Q&A, proactive surfacing, and teaching.	4 / 5
Workflow Clarity	A clear, sequenced retrieval workflow (outline.md → transcript.md range → safe excerpts → cite → fallback "say so explicitly") with a fallback checkpoint, but it is a simple linear process without the richer validation/error-recovery loop or checklist structure of a 5.	4 / 5
Progressive Disclosure	The structure is well designed with clearly signaled one-level-deep references to outline.md, transcript.md, and quote.md, but none of those referenced files are present in the bundle (no references/scripts/assets directories exist), so the disclosure design is unfulfilled and navigation leads to missing resources.	3 / 5
	Total	14 / 20 Passed

Description

90%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, highly distinctive description with excellent trigger-term coverage and a clear niche; its main weakness is that capability statements are embedded in a long topic enumeration rather than cleanly separated, leaving specificity and completeness slightly below their top anchors.

Dimension	Reasoning	Score
Specificity	The description names the domain richly but leans on enumerating talk topics ("the history of AI from the 1956 Dartmouth workshop", "training-data bias", "hobbyist AI tools") rather than distinct actions; the concrete capabilities boil down to answering questions, providing "safe excerpts", and "applying his framing to current AI work" — a few concrete actions, not comprehensive.	3 / 5
Completeness	Both what (answer questions, give safe excerpts, apply framing) and when ("Use when the user asks about Lieven Scheire's talk...") are present with concrete triggers, but the what is woven into the when-clause rather than stated as a crisp standalone capability, so it falls just short of a fully separated 5-anchor statement.	4 / 5
Trigger Term Quality	Comprehensive natural-term coverage including the speaker name, talk title, and many synonyms/examples users would say ("AI", "neural networks", "LLMs", "AGI", "pattern recognition", "black-box", plus distinctive phrases like "language imitation" and "snow in the background").	5 / 5
Distinctiveness Conflict Risk	Clear niche — one specific talk by one named speaker — with a highly distinctive primary trigger ("Lieven Scheire's talk 'Artificial Intelligence'") giving minimal conflict risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: jscraik/Agent-Skills
Path: Plugins/aidevcon/skills/talk-scheire-artificial-intelligence/SKILL.md
Commit: a3808a9

Reviewed: ~3 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.