AgentSkillsCN

project-feature-explainer

为项目功能提供专业讲解与说明。当您需要全面阐释某项功能的运作原理，包括功能概要、深度解析、使用示例以及序列/工作流程图时，可使用此技能。该技能的参考目录中包含额外的指导文件——`checklist.md`、`example-output.md` 与 `explanation-template.md`——每次分析都必须严格遵循这些文件。

SKILL.md
--- frontmatter
name: project-feature-explainer
description: Expert guidance for explaining project features. Use this skill when you need to provide a comprehensive explanation of how a specific feature works, including summaries, deep dives, usage examples, and sequence/workflow diagrams. This skill has references directory which contains additional instructions `checklist.md`, `example-output.md` and `explanation-template.md` that MUST be used for every analysis.

Project Feature Explainer Skill

This skill provides a systematic approach to analyzing and explaining a specific feature within a codebase.

Workflow

  1. •Identify Entry Points: Locate the main functions, classes, or API endpoints that trigger the feature.
  2. •Trace Dependencies: Identify the internal modules, services, or external APIs the feature relies on.
  3. •Analyze Data Flow: Understand how data enters the feature, how it's transformed, and where it's stored or returned.
  4. •Draft Explanation: Structure the explanation using the mandatory sections below.
  5. •Verify: Cross-reference the draft with the references/checklist.md to ensure completeness.

Mandatory Output Structure

1. Feature Summary

A high-level overview (1-2 paragraphs) explaining what the feature does and why it exists.

2. Deep Dive (Technical Details)

A detailed breakdown of the internal implementation.

  • •Key Components: List the main files/classes/functions involved.
  • •Logic Flow: Step-by-step description of the execution path.
  • •State Changes: Describe any side effects (e.g., database updates, cache invalidation).

3. Usage & Examples

Code snippets or CLI commands showing how to use the feature.

  • •Include common parameters and expected outputs.
  • •Provide a "Happy Path" example.

Guidelines

  • •Be Concise: Focus on the "how" and "why" without repeating obvious code logic.
  • •Reference Code: Use specific file paths and symbol names.
  • •Contextualize: Explain how this feature fits into the broader system architecture.