Token导航 LogoToken导航TokenDH.com
效率敏感数据clawhub未标认证来源可访问clear审计通过

openclaw-vertex-credit-safe-setupOpenClaw vertex credit safe 设置

Agent Skill

openclaw-vertex-credit-safe-setup 用于辅助部署、云资源、容器和基础设施运维,适合在 OpenClaw 中需要检查配置、整理部署步骤或排查环境问题时使用。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

4,618

周安装

185

GitHub Stars

公开资料未说明

下载量

1,495
OpenClaw

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

MIT-0

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:openclaw-vertex-credit-safe-setup(OpenClaw vertex credit safe 设置)
来源仓库:https://github.com/guoma970/openclaw-vertex-credit-safe-setup
安装命令:
openclaw skills install openclaw-vertex-credit-safe-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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install openclaw-vertex-credit-safe-setup

简介

使用 Google Cloud 项目、服务帐户 JSON 身份验证、最低成本验证和显式安全配置 Google Vertex AI,以实现全新的 OpenClaw 设置。

SKILL.md

name
openclaw-vertex-credit-safe-setup
description
Safely configure Google Vertex AI for a fresh OpenClaw setup using a Google Cloud project, service-account JSON auth, minimal-cost verification, and explicit billing checks to avoid accidental Gemini API or extra spend.
version
1.0.0

OpenClaw Vertex Credit-Safe Setup

Use this skill when a user wants to configure Google Vertex AI for OpenClaw from scratch and wants to minimize billing mistakes.

Goal

Set up Vertex AI in a way that:

  • uses a Google Cloud project
  • uses Vertex AI, not Gemini API direct access
  • prefers service-account JSON for first-time setup
  • keeps model routing on google-vertex/...
  • validates with one tiny request before wider use
  • explicitly checks billing / credits after the test

Use this skill when

  • the machine is a fresh or mostly fresh OpenClaw setup
  • the user wants to use Google free credits or trial credits safely
  • the user is unsure whether to use ADC, API keys, or service accounts
  • the user wants a repeatable first-time setup process
  • the user wants to avoid accidental extra spend from the wrong provider path

Required inputs

  • target Google Cloud project ID
  • confirmation that billing / credits are attached to that project
  • a local path where a service-account JSON file may be stored
  • permission to inspect current OpenClaw config files

Workflow

  1. Audit the current machine state.
  2. Confirm the target Google Cloud project and billing intent.
  3. Prefer a new service-account JSON path for first setup.
  4. Configure OpenClaw to use only google-vertex/... models for this setup.
  5. Run one minimal test request.
  6. Tell the user exactly what to check in Google Cloud Billing.
  7. Only after billing looks correct should broader rollout happen.

Guardrails

  • Do not present Gemini API direct setup as equivalent to Vertex AI setup.
  • Do not say setup is complete before the minimal test.
  • Do not say billing is correct unless the user checks the billing console.
  • Do not reuse ambiguous old auth state without stating it.
  • Do not broaden rollout across many agents until the tiny test is confirmed.

Preferred first-time auth path

For first-time setup, prefer:

  • Google Cloud project
  • Vertex AI API enabled
  • dedicated service account
  • dedicated service-account JSON file

Do not default to:

  • Gemini API key
  • mixed google/... and google-vertex/... models
  • unclear old ADC state
  • “it worked once on this machine” assumptions

Model routing rules

When configuring models, keep them on the Vertex provider path only:

  • google-vertex/gemini-3.1-pro-preview
  • google-vertex/gemini-3-pro-preview
  • google-vertex/gemini-2.5-pro
  • google-vertex/gemini-2.5-flash

Do not silently substitute:

  • google/...
  • Gemini API direct routes
  • partner-model assumptions when the user explicitly wants the safest likely credit-covered path

Required report format

Return:

  • current machine state
  • current Google path in use
  • target project
  • auth method
  • JSON path
  • configured google-vertex/... models
  • minimal test result
  • billing checks to perform next

Reference

For a compact checklist, read references/vertex-first-setup-checklist.md.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

98.64%
按下载量换算1,475

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

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。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