Zapier 效率

SKILL研究检索

原名 zapier

研究检索 接触敏感数据 许可证不明确 CodexClaudeCursorGemini CLI
安装这个技能

zapier 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。

累计安装 528 本周 22
GitHub 星标 55 仓库 vm0-skills
下载量 176 技能市场下载数
权限级别 接触敏感数据 安装前请确认风险范围
许可证 许可证不明确 unknown

安装

GitHub 仓库
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
npx skills add https://github.com/vm0-ai/vm0-skills --skill zapier

技能档案

研究检索
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方vm0-ai仓库 vm0-skills · 版本 未标注
宿主客户端Codex、Claude、Cursor、Gemini CLI技能可被这些客户端加载
安装方式GitHub 仓库安装包 vm0-ai/vm0-skills
权限级别接触敏感数据安装前请确认风险范围
许可证unknown许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验2026-05-01上游最后核验时间 · 数据更新 未标注
正文体量3.7 KBSKILL.md 原文大小,越大说明说明书写得越细

能做什么

4 项能力
上游给出的能力概览与适用场景
  • 按任务关键词查找相关 Skills
  • 展示可复制的安装命令
  • 保留来源站点、仓库和原始说明,方便继续核验
  • 展示第三方安全扫描或审计结果
用户想查找某类 Agent Skill 时需要根据任务场景推荐可安装能力包时需要对比不同来源的安装命令和来源信息时

权限与风险

接触敏感数据
装之前先读完这段

安装与使用指南

安装步骤

本站只展示命令,不托管安装包,也不自动执行。

怎么用

安装后应在对应宿主中按原始 README 的触发条件使用;当前草稿只做目录说明,具体调用方式由规则抽取,异常条目再复核。

SKILL.md 全文

3.7 KB
模型实际读到的说明书原文,来自技能仓库

Troubleshooting

If requests fail, run zero doctor check-connector --env-name ZAPIER_TOKEN or zero doctor check-connector --url https://actions.zapier.com/api/v2/check/ --method GET

How to Use

All examples below assume you have ZAPIER_TOKEN set. Authentication uses the x-api-key header.

1. Check API Key

Verify that your API key is valid.

curl -s "https://actions.zapier.com/api/v2/check/" --header "x-api-key: $ZAPIER_TOKEN" | jq .

2. List Exposed Actions

Retrieve all actions you have configured and exposed in your Zapier AI Actions dashboard.

curl -s "https://actions.zapier.com/api/v1/exposed/" --header "x-api-key: $ZAPIER_TOKEN" | jq '.results[] | {id, description, params}'

3. Execute an AI Action

Execute a configured action using natural language instructions. Replace ACTION_ID with the action ID from the list above.

curl -s -X POST "https://actions.zapier.com/api/v2/ai-actions/ACTION_ID/execute/" --header "Content-Type: application/json" --header "x-api-key: $ZAPIER_TOKEN" -d '{"instructions": "Send a message saying hello to the #general channel"}' | jq .

4. Execute with Parameter Hints

Supply parameter hints to guide the AI in filling action fields. Supported modes: locked (use exact value), guess (AI matches text), choose_from (AI selects from list), ignored (skip parameter).

Write to /tmp/zapier_request.json:

{
  "instructions": "Send an email about the weekly report",
  "params": {
    "to": {
      "mode": "locked",
      "value": "[email protected]"
    },
    "subject": {
      "mode": "locked",
      "value": "Weekly Report"
    }
  }
}

Then run:

curl -s -X POST "https://actions.zapier.com/api/v2/ai-actions/ACTION_ID/execute/" --header "Content-Type: application/json" --header "x-api-key: $ZAPIER_TOKEN" -d @/tmp/zapier_request.json | jq .

5. Preview an Action (Dry Run)

Preview what the action would do without actually executing it. Add preview_only=true as a query parameter.

curl -s -X POST "https://actions.zapier.com/api/v2/ai-actions/ACTION_ID/execute/?preview_only=true" --header "Content-Type: application/json" --header "x-api-key: $ZAPIER_TOKEN" -d '{"instructions": "Create a new row in the Sales spreadsheet with name John and amount 500"}' | jq .

6. Execute a V1 Exposed Action

Execute an action using the V1 endpoint. Replace ACTION_ID with the exposed action ID.

curl -s -X POST "https://actions.zapier.com/api/v1/dynamic/exposed/ACTION_ID/execute/" --header "Content-Type: application/json" --header "x-api-key: $ZAPIER_TOKEN" -d '{"instructions": "Send a Slack message to #dev saying deployment complete"}' | jq .

Guidelines

  1. Configure actions first: Before using the API, you must configure and enable actions in the Zapier AI Actions dashboard. The API can only execute actions you have set up
  2. Use natural language: The instructions field accepts plain English descriptions of what you want to do. The AI interprets and maps instructions to the correct action parameters
  3. Parameter modes: Use locked mode when you know the exact value, guess when the AI should match text to available options, and choose_from to provide a list of valid IDs
  4. Preview before executing: Use preview_only=true to verify the AI correctly interprets your instructions before running the action
  5. Response statuses: Check the status field in responses - values are success, error, empty, or preview
  6. Rate limits: Zapier enforces rate limits on API calls. Implement backoff if you receive HTTP 429 responses
  7. Action IDs: Each configured action has a unique ID. Use the list endpoint to discover available action IDs

安全信息

审计结论
未标注不代表已通过审计
安全审计Gen Agent Trust Hub、Socket、Snyk第三方安全审计平台给出的结论

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 2026-05-01。