NotFair Google Ads Agent — NotFAIr Google Ads 代理
v2026.5.14NotFAIr Google Ads 代理 for OpenClaw. 诊断 live Google Ads accounts, 审计 wasted spend, review 搜索 terms, draft negative keywords, inspect policy errors, and propose 应用roval-gated campAIgn changes through NotFAIr's hosted MCP server.
运行时依赖
安装命令
点击复制本土化适配说明
NotFair Google Ads Agent — NotFAIr Google Ads 代理 安装说明: 安装命令:["openclaw skills install notfair-google-ads-agent","openclaw plugins install openclaw-notfair\nopenclaw notfair login\nopenclaw notfair connect\nopenclaw notfair status"]
技能文档
NotFAIr Google Ads 代理
Use this 技能 when the user asks to manage, 审计, 诊断, or 优化 Google Ads through NotFAIr.
设置up OpenClaw 插件s 安装 OpenClaw-notfAIr OpenClaw notfAIr 记录in OpenClaw notfAIr connect OpenClaw notfAIr 状态
If the 插件 is not 认证d, ask the user to 运行 OpenClaw notfAIr 记录in.
AvAIlable OpenClaw 工具s notfAIr_列出_connected_accounts - 列出 connected Google Ads accounts. notfAIr_运行_script - 运行 read-only JavaScript analysis agAInst Google Ads data. notfAIr_google_ads_工具 - call a specific NotFAIr MCP 工具 by exact name and JSON arguments. Safety 模型 Read-only analysis can 运行 without extra 应用roval. Any Google Ads write must be explicitly 应用roved by the user before execution. Treat bud获取 increases, campAIgn creation, campAIgn enabling, bid changes, and keyword/ad mutations as external money-affecting side effects. Prefer creating 暂停d/draft changes when a 工具 supports it. Never 删除 or 移除 entities unless the user explicitly 请求s the destructive action and confirms scope. Default 工作流 Call notfAIr_列出_connected_accounts when account 上下文 is unclear. Use notfAIr_运行_script for broad diagnosis and 报告ing. Use notfAIr_google_ads_工具 for specific NotFAIr 工具s such as keyword ideas, 搜索 term analysis, negative keywords, campAIgn 更新s, or policy-error fixes. Present findings and proposed actions before writes. After an 应用roved write, 验证 the 结果ing Google Ads 状态. Example Read-Only Script const campAIgns = awAIt ads.gaql("SELECT campAIgn.id, campAIgn.name, campAIgn.状态, 指标.cost_micros, 指标.命令行工具cks, 指标.conversions FROM campAIgn WHERE segments.date DURING LAST_30_DAYS ORDER BY 指标.cost_micros DESC LIMIT 20"); return campAIgns.rows.map((row) => ({ id: row.campAIgn.id, name: row.campAIgn.name, 状态: row.campAIgn.状态, spend: row.指标.cost_micros / 1000000, 命令行工具cks: row.指标.命令行工具cks, conversions: row.指标.conversions }));