CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

scientific-slides

Builds slide decks and presentations for research talks. Used for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk. Provides slide structure, design templates, timing guidance, and visual validation. Works with PowerPoint and LaTeX Beamer.

Quality

79%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/scientific-slides/SKILL.md
SKILL.md
Quality
Evals
Security

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 delivers highly actionable, well-sequenced workflows backed by a genuine, correctly-linked bundle of reference and asset files. Its main weakness is redundancy — duplicated reference listings and near-verbatim command examples — plus deferred detail on the PptxGenJS build step.

Suggestions

Merge the two reference listings ("Core Capabilities" and "Reference Files") into a single annotated list so each bundled file appears exactly once.

Keep the full bash example set in one place only — replace the near-verbatim Quick Start command block with a pointer to the Image Workflow examples, or vice versa.

Trim motivational restatements ("Boring presentations = forgotten science", "Remember: Boring = Forgotten", and the 10-point Summary that repeats body content) to tighten the token budget.

DimensionReasoningScore

Conciseness

The body is mostly operational, but there is real padding: the "Reference Files" section re-lists five files (slide_capabilities, presentation_workflow, prompt_writing, script_reference, common_pitfalls) already introduced under "Core Capabilities"; the Quick Start Guide repeats the slide-generation bash commands nearly verbatim from the Image Workflow section; and motivational filler ("Boring presentations = forgotten science", "Remember: Boring = Forgotten", the 10-point Summary restating earlier content) could be cut without loss. This sits between the 'mostly efficient but could be tightened' (3) and 'noticeably verbose; several padded sections' (2) anchors — several distinct trimmable sections exist, but there is no over-explanation of concepts Claude already knows.

3 / 5

Actionability

Most guidance is executable: concrete copy-adaptable commands ("python scripts/generate_slide_image.py \"...\" -o slides/01_title.png", "python scripts/slides_to_pdf.py slides/*.png -o presentation.pdf"), an explicit prompt template, and verified real script/reference paths. It falls short of 5 due to minor gaps: the PPT workflow's "Step 2: Build PowerPoint with PptxGenJS" shows no code inline (deferred entirely to a bundled asset), and the schematics section points to an external "scientific-schematics skill documentation" that is not bundled.

4 / 5

Workflow Clarity

Both workflows (Image/PDF and PPT) are clearly sequenced (Plan → Generate → Combine/Build) with a formatting-consistency protocol, and the Quick Start includes an explicit feedback loop ("Open the PDF and review each slide / Regenerate any slides that need improvement / Re-combine") plus a validation pointer in Key Principles. It does not reach 5 because the Image Workflow's own three steps do not inline the validation checkpoint (it lives only in Quick Start and references/visual_review_workflow.md), and the bundled validate_presentation.py script is never invoked in the body.

4 / 5

Progressive Disclosure

The body is an overview that defers detail to real, one-level-deep bundled files — all 11 references/ and 5 assets/ files cited in the body exist on disk and match their annotations, with no nested reference chains. It falls short of 5 due to organization gaps: two overlapping reference listings ("Core Capabilities" plus a partially duplicative "Reference Files" section) and the duplicated inlined example-command blocks that belong in a single place.

4 / 5

Total

15

/

20

Passed

Description

83%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 with explicit 'what' and 'when' coverage and natural trigger terminology for the scientific-talk domain. Keyword coverage and action specificity are good but not exhaustive, leaving minor room to add synonyms like 'deck', '.pptx', or 'academic talk'.

DimensionReasoningScore

Specificity

The description lists several concrete capabilities — "Builds slide decks and presentations for research talks", "Provides slide structure, design templates, timing guidance, and visual validation", "Works with PowerPoint and LaTeX Beamer" — matching the anchor for several specific actions with minor gaps. It falls short of 5 because the core image-generation workflow and Beamer/PPTX authoring actions are only implied ('Works with'), not named as distinct actions, and above 3 because coverage goes well beyond 1-2 actions.

4 / 5

Completeness

Both questions are answered explicitly: what ("Builds slide decks and presentations for research talks... Provides slide structure, design templates, timing guidance, and visual validation") and when ("Used for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk") with concrete enumerated trigger phrases. This matches the 5 anchor; the 'when' clause is explicit and specific, not merely implied as at 4.

5 / 5

Trigger Term Quality

Natural user phrases are well covered: "PowerPoint slides", "conference presentations", "seminar talks", "research presentations", "thesis defense slides", "scientific talk", "LaTeX Beamer". A few common variations are missing (e.g., "slide deck", "defense slides", ".pptx", "academic talk"), which fits the 'good keyword coverage; a few natural terms missing' anchor rather than the comprehensive 5.

4 / 5

Distinctiveness Conflict Risk

The research-talk niche ("conference presentations", "seminar talks", "thesis defense slides", "LaTeX Beamer") is clearly staked out and mostly distinct. Minor overlap risk remains with a generic presentation/PowerPoint skill, since "PowerPoint slides" alone could trigger either — fitting the 'mostly distinct; minor overlap risk with closely related skills' anchor.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

metadata_field

'metadata' should map string keys to string values

Warning

Total

15

/

16

Passed

Repository

K-Dense-AI/scientific-agent-skills

Path

skills/scientific-slides/SKILL.md

Reviewed

Table of Contents

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.