Tessl • The package manager for agent skills

SkillPublic

kami-deck

A lab-meeting deck on gut-microbiome links to sleep quality — the design, the results, the caveats, and the next experiment. Built as a decision-grade academic research deck for lab group, PI.

Quality

65%

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 ./design-templates/kami-deck/SKILL.md

Quality

Content

67%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 well structured and token-efficient, with a genuine four-step workflow ending in an explicit self-check checklist and clear one-level-deep references. Its main weakness is actionability: the core build step relies entirely on mirroring external files (example.html, DESIGN.md, compose.ts) that are not part of this bundle, so no executable path to the deliverable exists from the skill file alone.

Suggestions

Inline a minimal index.html skeleton (head tokens, one slide of each kind, the nav runtime hook) so the 'Write index.html' step is executable even when example.html is not present in the bundle.

Turn the self-check into a feedback loop: after the checklist, add 'if any check fails, fix the slide and re-run the checklist before shipping'.

Consolidate the repeated ink-blue/parchment/no-italic rules into the DESIGN.md reference and keep only the slide-type table as the in-file source, trimming 'What you get' bullets that duplicate it.

Dimension	Reasoning	Score
Conciseness	The body is dense with skill-specific facts (tokens, hex values, slide types, keyboard model, ratios) and assumes Claude's competence without explaining known concepts. Not 5: the ink-blue/parchment and no-italic rules are repeated across 'What you get', the slide-type table, 'Self-check', and 'Boundaries', so some tokens could be consolidated into the DESIGN.md pointer.	4 / 5
Actionability	Concrete guidance exists — exact hex values, slide-type table, per-slide chrome spec, token source, and named files to mirror ('Mirror the structure of example.html', 'match the model documented in compose.ts') — but the core deliverable step delegates all structure and code to external files (example.html, DESIGN.md, compose.ts), none of which are included in this bundle. Not 4: without at least a minimal inline skeleton of index.html, the main instruction is not executable from what is written here.	3 / 5
Workflow Clarity	A clear four-step sequence (gather brief in two named rounds, pick language stack, write index.html, self-check) with an explicit validation checklist covering colors, numeric styling, navigation behavior, resize, and Lighthouse. Not 5: the self-check is a one-pass checklist with no fix-and-re-verify feedback loop, and step 3's structure is delegated to a file rather than specified.	4 / 5
Progressive Disclosure	The body is a well-organized overview with clearly signaled, one-level-deep references (DESIGN.md as token source of truth with section pointers, example.html, compose.ts, and a 'See also' list) rather than a monolith. Not 5: the referenced bundle files are not present alongside SKILL.md in this bundle — example.html in particular is instructed to be mirrored but does not exist here — so the split is structurally right but the referenced material is unavailable.	4 / 5
	Total	15 / 20 Passed

Description

48%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.

The description conveys a clear artifact concept and audience but reads as a description of one example deck rather than of the skill's capabilities. It lacks any 'use when' trigger guidance and concrete action verbs, which caps both completeness and specificity. Trigger terms are present but miss the natural synonyms users would say.

Suggestions

Add an explicit 'Use when...' clause naming the trigger situations (e.g., 'Use when asked to build lab-meeting slides, a group-meeting research presentation, or an academic review deck from a study or paper').

Replace the subject-specific framing with 2-3 concrete capability statements about what the skill does (e.g., 'Builds a single self-contained HTML slide deck from a study brief — gathers the brief in two rounds, writes the deck, and runs a design self-check').

Include natural trigger synonyms such as 'slides', 'presentation', 'group meeting', and 'deck' so retrieval works for phrasings users actually say.

Dimension	Reasoning	Score
Specificity	The description names the domain ('decision-grade academic research deck') and its content areas ('the design, the results, the caveats, and the next experiment') but describes the artifact rather than the skill's actions — the only action-adjacent phrase is 'Built as a... deck'. It does not reach anchor 3 because it never states 1-2 concrete capabilities the skill performs (e.g., producing an HTML slide deck, gathering a brief, running a critic pass).	2 / 5
Completeness	The 'what' is reasonably clear (a lab-meeting deck covering design, results, caveats, and next experiment), but the 'when' is entirely absent — 'for lab group, PI' names an audience, not a usage trigger, and there is no 'Use when...' clause or equivalent. Per the judging guideline, a missing 'Use when...' clause caps completeness at 3; it is not 2 because the 'what' is concrete, not vague.	3 / 5
Trigger Term Quality	Some natural terms are present — 'lab-meeting deck', 'academic research deck', 'research' — but common variations users would actually say ('slides', 'presentation', 'group meeting', file types) are missing, and the topic phrase 'gut-microbiome links to sleep quality' is an example subject rather than a trigger. Not 4: coverage lacks the synonyms and broader natural phrasings that would make retrieval reliable.	3 / 5
Distinctiveness Conflict Risk	The niche is fairly distinct ('lab-meeting', 'academic research deck', 'decision-grade' for lab group/PI), and the narrow subject framing reduces conflict with generic skills. It is not 5 because 'deck'/'research' overlap with closely related sibling deck skills (kami-landing, guizang-ppt, open-design-landing-deck referenced in the body), keeping minor overlap risk.	4 / 5
	Total	12 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 missing, 8 suspicious	Warning

	Total	14 / 16 Passed

Repository: nexu-io/open-design
Path: design-templates/kami-deck/SKILL.md
Revision: 78d5637

Reviewed: 17 days ago

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.

kami-deck

Table of Contents

Is this your skill?