仓库上下文

sodejm2026-10-25查看 391复制 1956

扫描和分析仓库结构以识别语言、框架、IaC、CI/CD 管道和凭据。

📘如何使用这个技能

1
复制技能内容·点击上方「Copy SKILL.md」复制全部内容，或 Download 下载到本地
2
打开你的 AI 客户端·支持 Claude Code、Trae、Cursor、ChatGPT、DeepSeek、Kimi 等任意 AI 助手
3
粘贴为系统提示/知识文件·Claude Code：放入 .claude/skills/ 目录；普通对话：作为首条消息发送并说「请严格按这份技能文档执行」
4
直接提出你的需求·AI 会按技能定义的工作流、格式和约束来处理，无需再解释背景

💡 提示：技能生效后，每次对话只需一句需求，AI 会按 SKILL.md 的流程执行。

SKILL.md1428 chars

---
name: repository-context
description: Scan and analyze repository structure to identify languages, frameworks, IaC, CI/CD pipelines, and credentials.
---

# Repository Context Collector Skill

This skill allows the agent to analyze a workspace or repository and gather critical architecture signals needed to customize security logging recommendations.

## Trigger Instructions

Run this skill during the initial stage of workspace analysis.

## Core Capabilities

1. **Deterministic Scan**: Invoke `scripts/collect-repository-context.py` to examine files in the workspace.
2. **Metadata Discovery**: Parse configuration manifests and package descriptors to identify:
   - Languages and frameworks (e.g., Node.js, Python, Go, Java)
   - Datastores (e.g., PostgreSQL, MongoDB, Redis)
   - Cloud providers (AWS, GCP, Azure) via IaC and configuration files
   - Identity & Authentication systems (JWT, OAuth, Keycloak, Auth0)
   - CI/CD pipelines (GitHub Actions, GitLab CI, Jenkins)
3. **Data Classification**: Identify sensitive entrypoints, public endpoints, and data flows to map data zones.
4. **Secrets Detection**: Check for credential patterns (e.g., API keys, passwords, database URLs). Report locations for technology context, but **never** copy or display the secret value.

## Output Generation

Save the output of this scan as a JSON/markdown payload matching the schema defined in `assets/repository-context.md`.

来源：skillsdirectory

仓库上下文

📘如何使用这个技能

评论0