What does this skill do, and when should you use it?
Golden_Html_Kw is a documentation-oriented Agent Skill for keyword categorization tests. It organizes four sections covering a getting-started guide, API usage, troubleshooting, and a changelog. The skill includes concepts, quick-reference mappings, code examples, tables, documentation statistics, and links to section reference files. It is useful for checking categorization and documentation-index behavior, but it is not presented as a general-purpose scraping or classification engine.
Provides a structured HTML documentation reference: it lists four sections and their categories, organizes topics such as Getting Started Guide, API Usage, Troubleshooting, and Changelog, shows Bash and Python examples, summarizes tables, code blocks, images, HTML files, and programming languages, and points to chapter files under references plus references/index.md. The supplied SKILL.md does not show executable scripts, command invocations, or network requests.
- A developer testing whether an agent recognizes golden_html_kw-related keywords can use the skill's topics and section labels as evaluation context.
- A tester checking document categorization into setup, API, troubleshooting, and changelog areas can use the Section Overview and Quick Reference.
- A developer validating code-example extraction can inspect the Bash installation example and the two Python examples.
- A documentation-index maintainer can use the Navigation section to locate section files such as references/section_s1-s1.md, references/section_s2-s2.md, and references/section_s3-s4.md.
How do you install this skill?
- Do not treat this test fixture as an executable documentation-conversion skill; it is primarily a fixed HTML output sample for keyword categorization.
- Before use, verify whether the long code example is complete and add concrete test assertions, failure feedback, source validation, and data-handling boundaries.
- The publisher is not verified in the FollowSkills enterprise registry, and the skill itself does not document Chinese-content handling or mainland-China network dependencies.
- Local filesystem
The repository is a monorepo containing 27 skills. The README provides this collection installation command:
pip install skill-seekers
This skill is located at tests/golden/phase2/html_kw/SKILL.md. The source does not document a standalone installation command for this file; to use it, provide the skill directory and its references files in a compatible Agent Skills directory.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/yusufkaraaslan/Skill_Seekers.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/tests/golden/phase2/html_kw" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Use golden-html-kw to check which documentation topics these keywords belong to
Enable the directory containing this SKILL.md in an Agent Skills-compatible client. For a keyword categorization test, a prompt such as “Use golden-html-kw to check which documentation topics these keywords belong to” can direct the agent to the section overview, key concepts, quick reference, and navigation. The source does not provide a dedicated CLI command or API for this individual skill.
What are this skill's strengths and limitations?
- Has a clearly defined purpose: keyword categorization testing.
- Combines section structure, topic indexing, code examples, table summaries, and documentation statistics.
- Includes navigation to section-level reference files.
- Uses only simple name and description frontmatter, with no visible platform-specific extensions.
- It is test-oriented and limited in scope, rather than a comprehensive production API knowledge base.
- The long Python example is truncated with an ellipsis in the supplied source and should not be treated as a complete runnable example.
- The Bash example uses the generic command pip install thing and is not a real project installation guide.
- The SKILL.md does not document standalone installation, a CLI entry point, network access, or an automatic categorization workflow.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Keyword Categorization Test Docs this page | 29 · Not recommended | ★ 15k | 25d ago | MIT |
| Golden EPUB Keyword Classifier | 38 · Not recommended | ★ 15k | 25d ago | MIT |
| Web Handbook Skill — Multi-Page Fixture | 41 · Not recommended | ★ 15k | 25d ago | MIT |
| Golden Word Documentation Skill | 30 · Not recommended | ★ 15k | 25d ago | MIT |
| EPUB Golden Build Guide | 36 · Not recommended | ★ 15k | 25d ago | MIT |
How did FollowSkills review this skill?
The skill only presents a documentation summary and reference files; it shows no credential access, external writes, destructive operations, or covert data flow, so no red-line risk is evident. It lacks source validation, sensitive-data handling, user confirmation, dependency-security, or rollback guidance, and the publisher is unverified, so points are deducted.
The Markdown structure is mostly coherent and includes an index, references, code blocks, and troubleshooting headings. However, the long Python example is truncated in SKILL.md, failure-handling content is empty, and there is no reproducible execution evidence for the skill's key path. Under static calibration, happy-path plausibility is limited, so the score remains below 10.
The trigger is clear but narrowly limited to keyword-categorization testing. The artifact represents a fixed HTML-document extraction result rather than the described capability of converting websites, GitHub repositories, and PDFs into Claude skills; non-fit boundaries and Chinese-language or mainland-China reachability are not stated, so substantial points are deducted.
The artifact has a name, trigger description, section overview, navigation, and reference files; repository context supplies an MIT license, version metadata, and CI. However, this generated test fixture lacks installation notes, stable parameters, changelog, maintenance ownership, update path, and limitation disclosure. Its troubleshooting and changelog sections are effectively empty, so only a mid-low score is justified.
As a golden fixture for keyword categorization, it offers inspectable section, code, table, and link structures. It cannot independently perform the described documentation/GitHub/PDF-to-Claude conversion, and its truncated example and placeholder-like content limit direct usability, so points are deducted.
The selected path includes an index and reference files, while repository context includes test configuration and a quality-metrics workflow, providing limited static traceability. There is no visible test assertion or CI evidence specifically covering this fixture, and no code was executed during review, so evidence coverage is thin and the static cap applies.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →