Agent Skills › ludo-technologies/pyscn › health-check

health-check

GitHub

使用pyscn对Python代码进行健康度检查，生成质量评分、等级及技术债务概览，支持前后对比与多维度指标分析。

.claude-plugin/pyscn-mcp/skills/health-check/SKILL.md ludo-technologies/pyscn

Trigger Scenarios

询问代码健康程度请求代码质量概览或评分评估重构后的质量变化查询技术债务情况

Install

npx skills add ludo-technologies/pyscn --skill health-check -g -y

More Options

Non-standard path

npx skills add https://github.com/ludo-technologies/pyscn/tree/main/.claude-plugin/pyscn-mcp/skills/health-check -g -y

Use without installing

npx skills use ludo-technologies/pyscn@health-check

指定 Agent (Claude Code)

npx skills add ludo-technologies/pyscn --skill health-check -a claude-code -g -y

安装 repo 全部 skill

npx skills add ludo-technologies/pyscn --all -g -y

预览 repo 内 skill

npx skills add ludo-technologies/pyscn --list

SKILL.md

Frontmatter

{
    "name": "health-check",
    "description": "Get an overall Python code quality health score using pyscn. Use when user asks how healthy or good the code is, wants a quality overview, a grade, a summary of technical debt, or a before\/after quality comparison."
}

Python Code Health Check with pyscn

Run the pyscn CLI to give a quick, quantified picture of code quality. No install needed:

uvx pyscn@latest analyze <path>

The summary output includes a health score (0-100), a letter grade (A-F), and per-category metrics.

Commands

User Request	Command
"How healthy is this code?"	`uvx pyscn@latest analyze <path>`
"Give me a quality overview"	Same command; walk through the category breakdown
"Did my refactoring improve quality?"	Run before and after, compare scores

For machine-readable detail add --json. Report files are NOT written to stdout. They go to .pyscn/reports/ under the current working directory and the path is printed on completion; read that file.

Interpreting Results

Score 0-100 with letter grade; category breakdowns cover complexity, dead code, duplication, coupling (CBO), cohesion (LCOM), dependencies, and architecture.
Lead with the grade and the weakest categories, then name the top offenders (files/functions) driving them.
For deeper follow-up, hand off to the focused skills: refactoring targets → refactoring, module structure → architecture-review.

Always explain the score in plain terms and suggest the highest-impact next step.

MCP Alternative

If the pyscn-mcp MCP server is connected, you can use its tools instead of the CLI: get_health_score (score + grade) or analyze_code (output_mode: "summary" gives the health score plus high-level metrics per category). Results are returned inline with no report files.