What does this skill do, and when should you use it?
hv-analysis is a research framework developed by Digital Life Khazix. It combines Saussure's diachronic-synchronic analysis, social science longitudinal-cross-sectional design, business school case study methods, and competitive strategy analysis. It guides the agent to collect information online, then perform a longitudinal (historical) and a cross-sectional (competitive) analysis, and finally synthesize insights across both axes, producing a polished PDF report (10,000-30,000 words). Ideal for systematic investigations, not for simple definitions.
The skill instructs agents to: 1. conduct parallel web searches (WebSearch/WebFetch, optional web-access skill, arXiv for academic topics); 2. create a longitudinal narrative tracing origins, evolution, key decisions, and phases (6,000-15,000 words); 3. identify competitors and compare core differences, user sentiment, ecosystem position, and trends (3,000-10,000 words); 4. cross-analyze to generate new insights and three future scenarios (1,500-3,000 words); 5. convert Markdown to PDF using its own script (scripts/md_to_pdf.py based on WeasyPrint).
- Product manager quickly researching a startup competitor to understand history and market position.
- Researcher systematically exploring a technical concept or research field, including related work and evolution.
- Investor evaluating a company, needing a deep report on founding background, funding, pivots, and competitors.
- Content creator gathering structured background before writing, to have a comprehensive brief.
- Analyst looking for a structured industry deep-dive, understanding key players and future directions.
How do you install this skill?
- Dependencies include WeasyPrint and Markdown; install command uses --break-system-packages, which may break system Python environment; users should assess before use.
- Skill heavily relies on overseas services (arXiv, GitHub, Reddit) that may be inaccessible from mainland China, affecting actual usage; ensure network reachability beforehand.
- Skill requires web searches and sub-agent calls, which involve many external requests; be mindful of privacy and data security; no explicit data handling policy.
- No test cases or failure handling are provided; if scripts or network fail, users may need to debug on their own, lacking official support.
- Shell / CLI
- Network access
Python 3weasyprintmarkdown
In Agent Skills-compatible clients like Claude Code or Codex, say: "Install this skill: https://github.com/KKKKhazix/khazix-skills/tree/main/hv-analysis". Alternatively, download the SKILL.md from the hv-analysis directory and place it in your project's skills folder (or follow Agent Skills standard installation).
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/KKKKhazix/khazix-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/hv-analysis" ~/.claude/skills/
rm -rf "$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 hv-analysis to research Hermes Agent
After installation, simply give the agent a subject, e.g., "Use hv-analysis to research Hermes Agent" or just "Research X". The skill will automatically search, analyze, and generate the PDF. Ensure Python dependencies are installed: pip install weasyprint markdown --break-system-packages. The final PDF is saved to the user's working directory as [Subject]_横纵分析报告.pdf.
What are this skill's strengths and limitations?
- Comprehensive, multi-disciplinary methodology with rigorous structure.
- Mandatory web search ensures rich information, with primary/secondary source priorities.
- Built-in PDF generation script produces a polished report.
- Adapts to many subject types (product, company, concept, person).
- Includes writing style guide, avoiding AI-flavored text for readability.
- Report generation is lengthy and requires extensive web searching.
- Heavily depends on agent web capabilities; limited without WebSearch support.
- Requires Python libraries like WeasyPrint, which complicates environment setup.
- Word count (10k+) may exceed some context lengths, risking truncation.
- No test suite or quality assurance; output quality varies with agent execution.
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 |
|---|---|---|---|---|
| hv-analysis: Deep Research Report this page | 46 · Use with care | ★ 21k | 10d ago | MIT |
| Exa Deep Research Orchestrator | 57 · Use with care | ★ 5.1k | 4d ago | MIT |
| Probability Analysis — Event Probability Analyst | 60 · Recommended | ★ 75 | 12d ago | MIT |
| Alpha Insights BizAdvisor | 59 · Recommended | ★ 67 | 3mo ago | MIT |
| Exa Agent Research Skill | 56 · Use with care | ★ 5.1k | 4d ago | MIT |
No direct competitor skills are mentioned in the source. It could be compared to generic deep-research tools, but no explicit comparison is provided.
How did FollowSkills review this skill?
The skill requires web searching, accessing arXiv API, running pip install, and executing a local script to convert Markdown to PDF. It involves external network calls and local script execution, but does not explicitly require user confirmation for each action; there is no mention of sensitive data handling or recovery/rollback mechanisms. No malicious behavior detected, but permissions and confirmation are incomplete.
The skill description is detailed, but the md_to_pdf.py script's dependencies (WeasyPrint, Markdown) require manual installation, and there is no failure handling or fallback. Static review cannot verify the key paths (web search and PDF generation) are runnable; no tests or error messages provided, so deductions apply.
The skill has clear use cases and trigger phrases, and is designed for Chinese users. However, it heavily depends on overseas services (arXiv, GitHub, Reddit) which may be unreachable from mainland China, reducing practical usability; while it is for deep research, it may be overkill for simple queries, but boundaries are well-defined.
Documentation is well-structured with methodology, steps, writing style, PDF generation instructions, and QA checklist, but lacks version history and changelog; installation note uses --break-system-packages which may affect system environment. Author identity and MIT license are clear, so overall convention is good.
The skill claims to produce 10k-30k word PDF reports, which is substantial, but static review cannot confirm output quality; the conversion script relies on external tools and may require debugging, so marginal value is unverified, hence not high scores.
Key claims like report quality and script usability lack executable tests or third-party verification; relies on external links (WeChat articles) which cannot be verified; static review cannot confirm, so deductions.
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 →