Token导航 LogoToken导航TokenDH.com
研究检索敏感数据github未标认证来源可访问许可证需确认审计异常

eve-new-project-setup前夕新项目设置

Agent Skill

eve-new-project-setup 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

6,217

周安装

241

GitHub Stars

公开资料未说明

下载量

1,909
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:eve-new-project-setup(前夕新项目设置)
来源仓库:https://github.com/incept5/eve-skillpacks
仓库路径:skills/eve-new-project-setup
安装命令:
npx skills add https://github.com/incept5/eve-skillpacks --skill eve-new-project-setup
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/incept5/eve-skillpacks --skill eve-new-project-setup

简介

eve-new-project-setup 处理 Eve Horizon 项目的初始化配置流程,包括档案创建和认证设置。

  • 适用于已执行 eve init 后仍需配置组织、项目和 git 远程地址的场景。
  • 提供 CLI 版本检查和 staging 环境档案创建的标准化步骤。
  • URL 结构由组织 slug 和项目 slug 决定,影响最终部署地址格式。
  • 安装前需确认是否有权限注册新组织和关联代码仓库。

SKILL.md

Eve New Project Setup

Use this after a developer has run eve init and needs to configure the project for Eve Horizon.

Context

The user has already run:

npm install -g @eve-horizon/cli
eve init my-project
cd my-project

This skill handles the remaining setup: profile, authentication, org/project registration, manifest customization, and git remote configuration.

Step 1: Verify CLI

eve --version

If this fails, the CLI wasn't installed. Have them run:

npm install -g @eve-horizon/cli

Step 2: Profile Setup

Create a profile for the staging environment:

eve profile create staging --api-url https://api.eh1.incept5.dev
eve profile use staging

Ask the user for their email and set defaults:

eve profile set --default-email user@example.com

Step 3: Authentication

Check current auth status:

eve auth status

If not authenticated, log in:

eve auth login

The CLI will guide them through SSH key discovery (can fetch from GitHub if needed).

We'll set harness credentials after project creation.

Step 4: Org + Project

Ask the user for:

  • Organization name (e.g., "my-company")
  • Project name (e.g., "My App")
  • Project slug (e.g., "my-app")
  • Repo URL (e.g., "git@github.com:me/my-app.git")

Create or ensure they exist:

eve org ensure my-company --slug myco
eve project ensure --name "My App" --slug my-app --repo-url git@github.com:me/my-app.git --branch main

URL impact: The org and project slugs directly form deployment URLs and K8s namespaces:

  • URL: {service}.{orgSlug}-{projectSlug}-{env}.{domain} (e.g., api.myco-my-app-staging.eh1.incept5.dev)
  • Namespace: eve-{orgSlug}-{projectSlug}-{env} (e.g., eve-myco-my-app-staging)
  • ${ORG_SLUG} is available for interpolation in manifest values

Slugs are immutable — choose short, meaningful values.

Set as defaults in the profile:

eve profile set --org org_xxx --project proj_xxx

Step 5: Choose Harness + Set API Keys

Ask which harness they want to target (can be more than one):

  • mclaude / claude (Anthropic)
  • code / codex (OpenAI)
  • zai (Z.ai)
  • gemini (Google)

Map harnesses to required secrets:

  • mclaude/claude: ANTHROPIC_API_KEY (preferred) or Claude OAuth via eve auth sync
  • code/codex: OPENAI_API_KEY (preferred) or CODEX_AUTH_JSON_B64
  • zai: Z_AI_API_KEY
  • gemini: GEMINI_API_KEY (or GOOGLE_API_KEY)

Decide where to store secrets:

  • Org (recommended) for shared keys across projects
  • Project for app-specific keys
  • User for personal keys

Use a secrets file for batch setup (starter repo includes secrets.env.example):

cp secrets.env.example secrets.env
eve secrets import --org org_xxx --file ./secrets.env

For Claude/Codex OAuth tokens on the host:

# Store OAuth tokens in the project scope (or add --system if admin)
eve auth sync --project proj_xxx

Step 6: Manifest Configuration

The starter template includes .eve/manifest.yaml. Update it with the project details:

schema: eve/compose/v2
project: my-app

services:
  api:
    build:
      context: apps/api
    ports: [3000]
    x-eve:
      ingress:
        public: true
        port: 3000

environments:
  staging:
    pipeline: deploy

pipelines:
  deploy:
    steps:
      - name: deploy
        action: { type: deploy }

Key fields to customize:

  • project: Match the project slug
  • services: Define your app's services
  • x-eve.ingress: Configure public access

Step 7: Git Remote

The project starts with no remote. Help set one up:

git remote -v
git remote add origin git@github.com:user/my-app.git
git push -u origin main

Step 8: Verification

Run these checks to confirm setup:

eve system health
eve auth status
eve profile show

Next Steps

After setup is complete, suggest:

  1. Run locally: docker compose up --build
  2. Set secrets: eve secrets set MY_KEY "value" or eve secrets import --file./secrets.env
  3. Deploy to staging: eve env deploy staging --ref main --repo-dir.
  4. Create a job: eve job create --prompt "Review the codebase"

Note: If the environment has a pipeline configured in the manifest, eve env deploy <env> --ref <sha> will trigger that pipeline. Use --direct to bypass the pipeline and do a direct deploy. --ref must be a 40-character SHA, or a ref resolved against --repo-dir/cwd.

Troubleshooting

"eve: command not found"

npm install -g @eve-horizon/cli

"Not authenticated"

eve auth login

"No profile"

eve profile create staging --api-url https://api.eh1.incept5.dev

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

35%
按下载量换算668

Claude

31.86%
按下载量换算608

Cursor

20.66%
按下载量换算394

Gemini CLI

9.02%
按下载量换算172

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills