CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

huashu-slides

从内容到成品PPTX的端到端演示文稿制作,含AI插画生成和18种设计风格。当用户提到"做PPT"、"做幻灯片"、"演示文稿"、"Keynote"、"slides"时使用。

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

73%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-engineered operational skill: the two-path workflow is unambiguous, defaults and checkpoints are explicit, and the AI-prompt guidance (Golden Rule, anti-pattern table, checklist) is genuinely non-obvious expertise. Its weaknesses are length-driven: the inline style catalog duplicates the gallery reference file and a promotional footer consumes tokens, and a few commands depend on external skills with placeholder paths.

Suggestions

Replace the 12 inline design-system preset specs with a short tiered summary table and point to references/proven-styles-gallery.md, which already contains the full 18-style details — this would cut ~100 duplicated lines.

Remove the promotional footer ("花叔出品 | 公众号 | 30万+粉丝...") and other non-operational branding from SKILL.md; it adds tokens without guiding execution.

Resolve the placeholder external paths: give a concrete way to locate or install the fallback script (e.g., a relative path to this bundle's own scripts/ directory) instead of "<本仓库路径>/huashu-wechat-image/scripts/generate_image.py" and ~/.claude/skills paths that are not distributed with the skill.

DimensionReasoningScore

Conciseness

Mostly dense and actionable (tables, commands, prompt templates), but ~100 lines of style presets (12 full inline descriptions with philosophy/colors/typography/composition) duplicate "references/proven-styles-gallery.md", and the promotional footer ("花叔出品 | 30万+粉丝...") is irrelevant padding. Not a 2 because little of the content explains concepts Claude already knows; not a 4 because the duplication and footer are clearly trimmable.

3 / 5

Actionability

Concrete executable guidance throughout: exact bash generation commands, html2pptx.js code snippet, create_slides.py invocation with flags, GOOD-vs-BAD prompt examples, and a prompt quality checklist. Minor gaps: the fallback uses a "<本仓库路径>" placeholder and several paths point to external skills (nano-banana-pro, pptx, image-to-slides) not distributed with this bundle.

4 / 5

Workflow Clarity

Explicit Step 0→5 sequence with mode/assembly-method defaults ("default to Guided", "default to Path A"), four labeled Checkpoints gated per collaboration mode, a post-generation quality checklist (text accuracy, layout, style consistency), and a feedback loop ("If a slide has text errors → regenerate with adjusted prompt").

5 / 5

Progressive Disclosure

References are clearly signaled, one level deep, and all exist in the bundle (design-movements.md, proven-styles-gallery.md, proven-styles-snoopy.md, prompt-templates.md, design-principles.md, assets/style-samples/). Docked one point because the 12 inline preset specs duplicate the gallery reference and the body also nests into another skill's files ("design-philosophy skill 的 references/design-styles.md").

4 / 5

Total

16

/

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: it states what the skill does with concrete specifics (AI illustration generation, 18 design styles, content-to-PPTX pipeline) and gives an explicit, natural 'use when' trigger list in both Chinese and English. The main gaps are a few missing common trigger variants (PowerPoint, .pptx) and no scoping to distinguish creation from presentation editing.

DimensionReasoningScore

Specificity

Lists several concrete capabilities — "从内容到成品PPTX的端到端演示文稿制作", "含AI插画生成", "18种设计风格" — though the umbrella phrase "端到端...制作" is broad and explicit actions (e.g., editing, converting) are not itemized.

4 / 5

Completeness

Explicitly answers both: what ("端到端演示文稿制作,含AI插画生成和18种设计风格") and when ("当用户提到...时使用") with concrete trigger phrases — matching the anchor-5 example pattern.

5 / 5

Trigger Term Quality

"做PPT", "做幻灯片", "演示文稿", "Keynote", "slides" are natural phrases users would say, with good CN/EN synonym coverage; misses common variants like "PowerPoint", ".pptx", and "deck/路演".

4 / 5

Distinctiveness Conflict Risk

A clear presentation-creation niche with distinct triggers (做PPT/Keynote/slides); minor overlap risk with a general pptx-editing skill since "slides"/"演示文稿" could also arise for editing tasks, which the description does not scope out.

4 / 5

Total

17

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (642 lines); consider splitting into references/ and linking

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

14

/

16

Passed

Repository

alchaincyf/huashu-skills

Path

huashu-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.