CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

tessleng/aidevcon-slides

Build AI DevCon NYC talk decks as Claude Slides in the official aiDevCon style, with an HTML fallback.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

Quality

Content

92%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.

The body is an exemplary dense, actionable workflow: concrete parameters, a complete JSON template, explicit validation and fallback branches, and well-organized one-level references. The one material defect is bundle integrity — the `templates/` directory the whole copy-the-layout workflow (and build_deck.py) depends on is absent, which blocks the skill's core instruction and caps progressive disclosure.

Suggestions

Ship the referenced `templates/layouts/*.html` (24 slide layouts) and `templates/deck.html` in the bundle — the body's core instruction ("Copy, fill, never redesign") and scripts/build_deck.py lines 50-53 read from `templates/`, but no such directory exists, so the primary workflow cannot be executed.

If the templates are intentionally hosted elsewhere (e.g. inside the Slides design system artifact), say so explicitly in the body's "What ships with this skill" list so the reader knows where to obtain the layouts.

State the fix-and-rerun loop for the pre-publish checklist inline (e.g. "fix every failure and re-run it until clean") rather than relying on the reader opening references/checklist.md to learn that validation is iterative.

DimensionReasoningScore

Conciseness

Lean and dense throughout: no concept re-teaching, and every sentence carries executable specifics ("The layouts use 64px side and 72px top and bottom margins, and the aiDevCon wordmark as the footer at bottom:40px"). Short rationale clauses ("so the conference feels like one event") are behavioral guardrails, not padding — nothing to trim.

5 / 5

Actionability

Fully copy-paste ready: a complete `deck.json` JSON literal with real font hrefs and the design-system artifact URL, exact publish call parameters ("action: \"publish\", the deck's url, asset: true, file_paths"), exact commands ("python3 scripts/build_deck.py deck.html"), and a defined placeholder format ("[Placeholder: % of findings caught before review]").

5 / 5

Workflow Clarity

A 7-step numbered sequence with an explicit validation checkpoint ("Run references/checklist.md on the files" before publish, and the checklist itself mandates fixing every failure) plus error-recovery branches: the design system being unreadable ("leave both out and say so in one line") and the no-Slides-artifact fallback ending in a screenshot check via check_deck.py.

5 / 5

Progressive Disclosure

References are well signaled and one level deep (layouts.md, design-rules.md, checklist.md, scripts), but the bundle is missing its central referenced component: no `templates/` directory ships, even though the body's core instruction ("templates/layouts/*.html: 24 slides… Copy, fill, never redesign") and scripts/build_deck.py both depend on templates/layouts/*.html and templates/deck.html. A missing, load-bearing referenced path is a structural gap beyond the "minor organization gaps" of the next anchor up.

3 / 5

Total

18

/

20

Passed

Description

87%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 description: concrete what, explicit multi-trigger when-clause in third-person imperative voice, and a tightly scoped niche that keeps conflict risk low. Keyword coverage is good but misses a few natural synonyms like "presentation" or "speaker deck".

DimensionReasoningScore

Specificity

Lists several concrete actions ("create, draft, restyle or fix slides", "convert an outline or an existing deck into the DevCon template", "add slides") with named style specifics ("black, yellow, pixel art, Atkinson Hyperlegible"), but capabilities like editing or exporting are only implied — minor gaps in coverage, not comprehensive.

4 / 5

Completeness

Clearly answers both: the what ("Build an AI DevCon NYC talk deck as a Claude Slides artifact in the official aiDevCon style") and an explicit when with concrete trigger phrases ("Use when a speaker asks to create, draft, restyle or fix slides for an AI DevCon talk or workshop, convert an outline…").

5 / 5

Trigger Term Quality

Good natural-keyword coverage: "slides", "talk or workshop", "outline", "existing deck", "DevCon template", "restyle", "fix slides" — phrases a speaker would naturally say. A few common synonyms (e.g. "presentation", "speaker deck") are missing, so it falls just short of comprehensive.

4 / 5

Distinctiveness Conflict Risk

A clear niche ("AI DevCon NYC", "DevCon template") with triggers that name it explicitly; generic slide-making requests would not fire this skill, so conflict risk is minimal.

5 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

On this page