CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

paper-slides

Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script. Use when user says "做PPT", "做幻灯片", "make slides", "conference talk", "presentation slides", "生成slides", "写演讲稿", or wants beamer slides for a conference talk.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/skills-codex-gemini-review/paper-slides/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

70%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 a well-sequenced, validation-rich 9-phase workflow with concrete commands and a ready-to-use beamer template. Its weaknesses are redundancy (key presentation rules stated twice) and a monolithic inline structure — long templates that should live in reference files, plus a PPTX export step specified only as instructions.

Suggestions

Move the full TALK_SCRIPT.md template (including the 8-question Q&A block), the per-talk-type slide tables, and the beamer template into reference files (e.g., references/talk-script-template.md, references/slide-templates.md) linked from the body, keeping SKILL.md as an overview.

Deduplicate the presentation rules: keep the Phase 2 rules table and have Key Rules reference it instead of restating 'one message per slide', '≥60% figure area', and font minimums verbatim.

Bundle an executable generate_pptx.py script (or a references/ implementation guide with runnable code) instead of describing its behavior as a spec Claude must implement from scratch.

DimensionReasoningScore

Conciseness

Mostly efficient tables and commands, but there is real tightening material: 'One message per slide', 'figure ≥60% of slide area', font-size minimums, and progressive-disclosure rules each appear twice (Phase 2 rules table AND Key Rules); slide-count guidance is given in Constants and again per talk type; and the ~90-line TALK_SCRIPT/Q&A template pads the body. Not level 4 because the duplication exceeds 'minor instances'.

3 / 5

Actionability

Copy-paste-ready beamer template, concrete bash commands (latexmk, pdfinfo, symlink commands), named MCP tools, and an explicit compile error-retry loop — but Phase 7 specifies generate_pptx.py as requirements for Claude to write rather than providing executable code or a bundled script, leaving a minor gap versus the fully executable level-5 anchor.

4 / 5

Workflow Clarity

Phases 0-8 are explicitly sequenced with state persisted to SLIDES_STATE.json after each phase, an explicit STOP checkpoint with enumerated user options, a bounded compile-retry loop (max 3 attempts), pdfinfo page-count verification, and startup resume logic — a clear feedback-rich sequence.

5 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent), and long inline material that belongs in reference files is monolithically embedded — the full TALK_SCRIPT template with 8 Q&A examples, the beamer template, and per-talk-type slide tables. The three shared-protocol links point outside the bundle ('../../shared-references/*.md') with unverifiable paths. Structure exists but content that should be split out is inline, matching level 3.

3 / 5

Total

15

/

20

Passed

Description

88%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 that states concrete deliverables (beamer PDF, editable PPTX, speaker notes, full talk script) and gives an explicit, bilingual 'Use when' trigger list. The only gaps are a few missing natural trigger variants ('PPT', 'deck', file extensions) and slight overlap risk with generic presentation skills.

Suggestions

Add common English trigger variants such as 'PPT', 'slide deck', 'deck', and file extensions (.pptx, .tex) alongside the existing bilingual trigger list.

Narrow the generic 'presentation slides' trigger toward the paper-derived context (e.g., 'paper slides', 'talk for my paper') to reduce overlap with a general pptx-editing skill.

DimensionReasoningScore

Specificity

Enumerates multiple concrete, specific actions — 'beamer LaTeX → PDF + editable PPTX', 'speaker notes', 'full talk script' — covering the full deliverable set comprehensively, matching the level-5 anchor rather than the 'minor gaps' of level 4.

5 / 5

Completeness

Explicitly answers both questions: the 'what' ('Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script') and the 'when' ('Use when user says ... or wants beamer slides for a conference talk') with concrete quoted trigger phrases.

5 / 5

Trigger Term Quality

Good natural coverage including bilingual synonyms ('做PPT', '做幻灯片', 'make slides', 'conference talk', 'presentation slides', '写演讲稿') but misses common English variations users would say — 'PPT', 'deck'/'slide deck', and file extensions (.pptx, .tex) — so it falls short of the comprehensive level-5 anchor.

4 / 5

Distinctiveness Conflict Risk

Clear niche (conference talk slides generated from a compiled paper) with specific triggers, but generic phrases like 'make slides' and 'presentation slides' could also fire for a general pptx-presentation or poster skill, giving minor overlap risk with closely related skills.

4 / 5

Total

18

/

20

Passed

Validation

75%

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

Validation — 12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (583 lines); consider splitting into references/ and linking

Warning

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 3 suspicious

Warning

Total

12

/

16

Passed

Repository

wanshuiyin/Auto-claude-code-research-in-sleep

Path

skills/skills-codex-gemini-review/paper-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.