后端项目接手盘点指南
安装方式
先用 Prompt 核对来源,也可切换到安装命令或下载可用的安装包。
用 AI 助手安装
复制提示词,让助手先核对固定版本和依赖。
查看 Prompt收起 Prompt
先检查当前 Agent 是否已安装 SkillPI Skill Finder。如果尚未安装,请按官方仓库的说明安装并启用:https://github.com/Findskills/skillpi。然后使用 Skill Finder 打开此 SkillPI 详情页并定位 "spine-recon":https://skillpi.com/zh/skills/deepskill-31bb7a1071e1225c9725a5f72700b9d3c75ece0bbed7093384231e89e551822d。安装前核对上游来源和版本:https://github.com/jeremylongshore/tons-of-skills-marketplace/blob/6ea46b08dce892944db7c16f28d0b5308d402402/plugins/ai-agency/tonone/skills/spine-recon/SKILL.md。来源不一致时停止并说明差异。核对通过后,用 Skill Finder 安装目标 Skill,并报告安装路径和来源版本。
此命令从 GitHub 来源仓库获取指定技能。运行前请先查看源文件。
npx skills add https://github.com/jeremylongshore/tons-of-skills-marketplace/tree/6ea46b08dce892944db7c16f28d0b5308d402402/plugins/ai-agency/tonone/skills/spine-recon --skill spine-recon来源与配置
查看 GitHub 源码jeremylongshore/tons-of-skills-marketplace固定提交 6ea46b08dce892944db7c16f28d0b5308d402402
配置方式待核实
请查看原始来源中的适用工具、依赖和权限要求。
安装实测待补充
按框架、依赖、路由、中间件、数据模型、外部服务和认证机制梳理后端项目,并形成质量与技术债务清单。适合获授权的代码库交接或初步评估;扫描前确认代码与数据访问范围,输出中可能涉及内部端点或架构信息,应先脱敏再共享。
文件浏览
--- name: spine-recon description: Backend reconnaissance — map all routes, middleware, models, dependencies, auth, and assess code quality for project takeover. Use when asked to "understand this backend", "map the API", or "assess code quality". allowed-tools: Read, Bash, Glob, Grep, WebFetch, WebSearch, AskUserQuestion version: 0.6.4 author: tonone-ai <hello@tonone.ai> license: MIT --- # Backend Reconnaissance You are Spine — the backend engineer from the Engineering Team. Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose. ## Steps ### Step 0: Detect Environment ```bash ls -a ``` Identify the framework, language, package manager, database, and infrastructure. Read package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, or Gemfile for the full dependency list. ### Step 1: Map All Routes and Endpoints Find and read all route definitions. Build a complete endpoint map: | Method | Path | Auth | Handler | Description | | ------ | ---------- | ---- | --------------------- | ----------- | | GET | /api/users | JWT | UserController.list | List users | | POST | /api/users | JWT | UserController.create | Create user | Note any undocumented endpoints, debug routes, or admin endpoints. ### Step 2: Map Middleware Stack Identify the middleware execution order: 1. Request logging 2. CORS 3. Auth (JWT / API key / session) 4. Rate limiting 5. Body parsing / validation 6. Route handler 7. Error handling Note any middleware that applies globally vs. per-route. ### Step 3: Map Database Models List all database models/tables with: - Fields and types - Relationships (foreign keys, many-to-many) - Indexes - Migrations status (up to date, pending) ### Step 4: Map External Dependencies Identify all external services the backend calls: - Third-party APIs (payment, email, auth providers) - Cloud services (S3, Pub/Sub, SQS) - Other internal services For each: note the client library used, timeout configuration, and circuit breaker status. ### Step 5: Assess Auth Mechanism Document: - Auth type (JWT, session, API key, OAuth2, mTLS) - Token storage and validation approach - Role/permission model - Which endpoints are public vs. protected ### Step 6: Assess Code Quality Evaluate: - **Test coverage** — are there tests? What percentage of routes are tested? - **Code quality signals** — consistent naming, clear separation of concerns, no god files - **Tech debt hotspots** — large files (>500 lines), TODOs/FIXMEs, commented-out code, complex functions - **Error handling** — consistent patterns or ad-hoc try/catch everywhere? - **Dependency freshness** — are dependencies up to date or significantly behind? - **Documentation** — API docs, README, inline comments on complex logic ### Step 7: Present the Assessment Format as: ``` ## Backend Recon: [project name] **Stack:** [language] + [framework] + [database] **Routes:** [X] endpoints across [Y] resources **Test coverage:** [estimated percentage or "none"] ### Route Map [endpoint table from Step 1] ### Architecture - **Auth:** [mechanism] - **Middleware:** [stack summary] - **Database:** [X] models, [Y] migrations - **External deps:** [list with timeout/circuit breaker status] ### Code Quality | Signal | Status | Notes | |-------------------|---------------|------------------------------| | Test coverage | Low/Med/High | [details] | | Error handling | Consistent/Ad-hoc | [details] | | Dependency health | Current/Stale | [X deps behind major versions] | | Tech debt | Low/Med/High | [hotspot files] | ### Takeover Recommendations 1. [First thing to do when taking over this codebase] 2. [Second priority] 3. [Third priority] ``` Map for someone inheriting the project. Factual, specific, actionable. ## Delivery If output exceeds the 40-line CLI budget, invoke `/atlas-report` with the full findings. The HTML report is the output. CLI is the receipt — box header, one-line verdict, top 3 findings, and the report path. Never dump analysis to CLI.
当前收录版本
0.6.4
质量评测
满分 100。前四项来自 DeepSkill 收录的质量报告。
评分依据
提供后端项目接管的端到端侦察能力,覆盖路由、中间件、模型、依赖、认证和代码质量评估
评分依据
触发描述精准,明确覆盖“理解后端”、“映射 API”、“评估代码质量”等典型场景
评分依据
步骤划分清晰,共 7 个逻辑阶段,每个阶段有明确目标和输出格式
评分依据
工具使用合理,均为平台内置工具,参数正确
计算方式
取前四项原始分值的算术平均,四舍五入取整。可通过收录报告核对各项评分与说明,运行记录需单独查证。
权限与使用前检查
根据当前主文档自动整理,命中内容可能是示例或防护说明;不是实际权限声明,也未覆盖附件、外部依赖或运行行为。
-
账号、密钥或授权
使用专用、最小权限账号,不把密钥粘贴到公开内容或命令日志;核对接收数据的服务与租户。
主文档提及 · 第 43, 74 行
安全评级与访问范围
risk · 静态证据核验,不代表运行安全保证。
使用前请由宿主核对并授权实际文件读写、外部服务和联网目的地;本站的核验不会授予任何运行权限。
用户讨论
还没有已核实的点评。