Install any skill in seconds. Free to start, no credit card required.
Get Started Free →This skill should be used when the user asks about "mdBook", "mdbook", "book.toml", "SUMMARY.md", "mdbook build", "mdbook serve", "mdbook init", "mdbook watch", "mdbook test", "mdbook clean", "mdBook configuration", "mdBook preprocessor", "mdBook renderer", "mdBook backend", "mdBook theme", "mdBook syntax highlighting", "mdBook MathJax", "mdBook search", "mdBook CI deployment", "mdBook GitHub Pages", "mdBook GitLab Pages", "mdBook custom theme", "index.hbs", "mdBook editor", "mdBook markdown", "
.claude/skills/marceloeatworld-mdbook/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 17 |
| gemini-3.1-pro-preview | 100% | 3 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | -82% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-22 | ✓→✓ | = Same ✓ | 32% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 80% | 0% |
Complete reference for mdBook, the Rust-based tool for creating books from Markdown. Auto-generated from the official guide at https://github.com/rust-lang/mdBook.
The references/ directory contains the full, unmodified guide markdown, updated daily.
book.toml (TOML) syntax for configuration examples.SUMMARY.md format for table of contents structure.Identify the topic from the user's question, then read the matching reference file:
| Topic | File | |-------|------| | Overview and introduction | references/overview.md | | Installation (cargo, binaries) | references/installation.md | | Creating a new book | references/creating-a-book.md | | Reading / navigating books | references/reading-books.md |
| Topic | File | |-------|------| | All CLI commands (init, build, watch, serve, test, clean, completions) | references/cli.md |
| Topic | File | |-------|------| | SUMMARY.md structure and syntax | references/summary-format.md | | Markdown features and extensions | references/markdown.md | | mdBook-specific features (hiding code, include, playground) | references/mdbook-specific.md | | MathJax support | references/mathjax.md |
| Topic | File | |-------|------| | book.toml (general, preprocessors, renderers, environment variables) | references/configuration.md |
| Topic | File | |-------|------| | Theme customization (index.hbs, syntax highlighting, editor) | references/theme.md |
| Topic | File | |-------|------| | Writing preprocessors and backends (Rust API) | references/for-developers.md |
| Topic | File | |-------|------| | CI/CD (GitHub Actions, GitLab CI) | references/continuous-integration.md |
When reference files are insufficient, fetch the latest docs from raw GitHub:
https://raw.githubusercontent.com/rust-lang/mdBook/main/guide/src/<path>.mdExamples:
https://raw.githubusercontent.com/rust-lang/mdBook/main/guide/src/format/configuration/general.mdhttps://raw.githubusercontent.com/rust-lang/mdBook/main/guide/src/cli/build.mdhttps://raw.githubusercontent.com/rust-lang/mdBook/main/guide/src/format/theme/syntax-highlighting.mdbook.toml and Markdown for SUMMARY.md.references/for-developers.md.references/continuous-integration.md.bashmdbook init my-book cd my-book mdbook serve --open # preview at http://localhost:3000
my-book/
├── book.toml # Configuration
├── src/
│ ├── SUMMARY.md # Table of contents
│ ├── chapter_1.md
│ └── chapter_2/
│ ├── section_1.md
│ └── section_2.md
└── book/ # Generated output (after build)markdown# Summary [Introduction](README.md) - [Chapter 1](chapter_1.md) - [Section 1.1](chapter_1/section_1.md) - [Chapter 2](chapter_2.md) --- [Appendix](appendix.md)
toml[book] title = "My Book" authors = ["Author Name"] language = "en" src = "src" [build] build-dir = "book" [output.html] default-theme = "light" preferred-dark-theme = "ayu" git-repository-url = "https://github.com/user/repo"
bashmdbook build # build to book/ mdbook serve # live preview server mdbook watch # rebuild on changes mdbook test # test Rust code samples mdbook clean # remove build artifacts
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | pass→pass | 15,924 | 18,765 | +18% | 1 | 1 | 0% | 1,700 | 2,241 | +32% | 0 | 0 | — |
case-01 | pass→pass | 14,653 | 13,738 | -6% | 1 | 1 | 0% | 1,599 | 2,886 | +80% | 0 | 0 | — |
case-02 | pass→pass | 12,575 | 10,545 | -16% | 1 | 1 | 0% | 1,262 | 2,158 | +71% | 0 | 0 | — |
case-03 | pass→pass | 17,133 | 14,725 | -14% | 1 | 1 | 0% | 2,270 | 2,878 | +27% | 0 | 0 | — |
case-04 | pass→pass | 10,524 | 19,603 | +86% | 1 | 1 | 0% | 968 | 2,216 | +129% | 0 | 0 | — |
case-05 | pass→pass | 11,316 | 9,260 | -18% | 1 | 1 | 0% | 1,038 | 1,837 | +77% | 0 | 0 | — |
case-06 | pass→pass | 10,300 | 8,317 | -19% | 1 | 1 | 0% | 712 | 1,678 | +136% | 0 | 0 | — |
case-07 | pass→pass | 10,565 | 8,727 | -17% | 1 | 1 | 0% | 832 | 1,764 | +112% | 0 | 0 | — |
case-08 | pass→pass | 10,203 | 8,980 | -12% | 1 | 1 | 0% | 870 | 1,849 | +113% | 0 | 0 | — |
case-09 | pass→pass | 12,403 | 9,550 | -23% | 1 | 1 | 0% | 1,089 | 1,953 | +79% | 0 | 0 | — |
case-10 | pass→pass | 7,105 | 15,994 | +125% | 1 | 1 | 0% | 1,134 | 1,738 | +53% | 0 | 0 | — |
case-11 | pass→pass | 7,633 | 8,179 | +7% | 1 | 1 | 0% | 299 | 1,596 | +434% | 0 | 0 | — |
case-16 | pass→pass | 11,144 | 6,596 | -41% | 1 | 1 | 0% | 1,008 | 1,459 | +45% | 0 | 0 | — |
case-12 | pass→pass | 7,653 | 8,521 | +11% | 1 | 1 | 0% | 410 | 1,692 | +313% | 0 | 0 | — |
case-13 | pass→pass | 7,976 | 8,174 | +2% | 1 | 1 | 0% | 480 | 1,717 | +258% | 0 | 0 | — |
case-14 | pass→pass | 7,474 | 8,035 | +8% | 1 | 1 | 0% | 428 | 1,737 | +306% | 0 | 0 | — |
case-15 | pass→pass | 8,568 | 7,407 | -14% | 1 | 1 | 0% | 473 | 1,605 | +239% | 0 | 0 | — |
case-17 | fail→pass | 53,600 | 7,523 | -86% | 1 | 1 | 0% | 8,796 | 1,610 | -82% | 0 | 0 | — |
case-18 | fail→pass | 17,464 | 7,767 | -56% | 1 | 1 | 0% | 2,036 | 1,626 | -20% | 0 | 0 | — |
case-19 | fail→pass | 14,015 | 7,466 | -47% | 1 | 1 | 0% | 1,366 | 1,534 | +12% | 0 | 0 | — |
case-20 | pass→pass | 12,072 | 7,935 | -34% | 1 | 1 | 0% | 1,346 | 1,759 | +31% | 0 | 0 | — |
case-21 | pass→pass | 20,570 | 18,882 | -8% | 1 | 1 | 0% | 2,690 | 2,124 | -21% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/6/2026 | +27% |
Other measured skills in the registry, with their headline benchmark lift.