Token导航 LogoToken导航

Agent Skill · 研究检索

apideck-node(顶层甲板节点)

apideck-node 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。

apideck-node 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为接触敏感数据;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 64,379。

总安装 303周安装 13GitHub Stars 2下载 106核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:apideck-node(顶层甲板节点)
来源仓库:https://github.com/apideck-libraries/api-skills
仓库路径:skills/apideck-node
安装命令:
npx skills add https://github.com/apideck-libraries/api-skills --skill apideck-node
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/apideck-libraries/api-skills --skill apideck-node

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

TypeScript/JavaScript 对接 Apideck 的统一 API SDK。通过 npm 安装 @apideck/unify 获得完整类型定义。支持会计、CRM、HRIS、文件存储等多领域连接器。所有请求均需携带认证头与指定服务标识符。

apideck-node 属于研究检索类 Skill,可作为该场景下的辅助能力补充。推荐在服务端使用此 SDK 确保密钥安全。

来源仓库内路径:skills/apideck-node

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/apideck-libraries/api-skills --skill apideck-n…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

apideck-node 怎么安装?

在终端执行 npx skills add https://github.com/apideck-libraries/api-skills --skill apideck-node。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

apideck-node 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

apideck-node 需要什么权限?

权限级别是接触敏感数据。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。

apideck-node 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

apideck-node 用的人多吗?

公开数据是累计安装 303 次,近一周安装 13 次,下载 106 次,GitHub 2 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

apideck-node 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
apideck-node

顶层甲板节点 · 本页

3032接触敏感数据未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Apideck TypeScript SDK Skill

Overview

The Apideck Unified API provides a single integration layer to connect with 200+ third-party services across accounting, CRM, HRIS, file storage, ATS, e-commerce, and more. The official TypeScript SDK (@apideck/unify) provides typed clients for all unified APIs.

Key capabilities:

  • Accounting - Invoices, bills, payments, ledger accounts, journal entries, tax rates, balance sheets, P&L
  • CRM - Contacts, companies, leads, opportunities, activities, pipelines, notes
  • HRIS - Employees, departments, payrolls, time-off requests, schedules
  • File Storage - Files, folders, drives, shared links, upload sessions
  • ATS - Jobs, applicants, applications
  • Vault - Connection management, OAuth flows, custom field mapping
  • Vault JS - Embeddable modal UI for users to authorize connectors and manage settings
  • Webhook - Event subscriptions and real-time notifications

Installation

npm add @apideck/unify

Requires Node.js 18+. The SDK is fully typed with TypeScript definitions.

IMPORTANT RULES

  • ALWAYS use the @apideck/unify SDK. DO NOT make raw fetch/axios calls to the Apideck API.
  • ALWAYS pass apiKey, appId, and consumerId when initializing the client. These are required for all API calls.
  • ALWAYS set the APIDECK_API_KEY environment variable rather than hardcoding API keys.
  • USE serviceId to specify which downstream connector to use (e.g., "salesforce", "quickbooks", "xero"). If a consumer has multiple connections for an API, serviceId is required.
  • USE cursor-based pagination with for await...of for iterating large result sets. DO NOT implement manual pagination.
  • USE the filter parameter to narrow results server-side. DO NOT fetch all records and filter client-side.
  • USE the fields parameter to request only the columns you need. This reduces response size and improves performance.
  • ALWAYS handle errors with try/catch. The SDK throws typed errors for different HTTP status codes.
  • DO NOT store Apideck API keys, App IDs, or Consumer IDs in source code. Use environment variables or a secrets manager.

Quick Start

import { Apideck } from "@apideck/unify";

const apideck = new Apideck({
  apiKey: process.env["APIDECK_API_KEY"] ?? "",
  appId: "your-app-id",
  consumerId: "your-consumer-id",
});

// List CRM contacts
const { data } = await apideck.crm.contacts.list({
  limit: 20,
  filter: { email: "john@example.com" },
});

for (const contact of data) {
  console.log(contact.name, contact.emails);
}

SDK Patterns

Client Setup

import { Apideck } from "@apideck/unify";

const apideck = new Apideck({
  apiKey: process.env["APIDECK_API_KEY"] ?? "",
  appId: "your-app-id",
  consumerId: "your-consumer-id",
});

The consumerId identifies the end-user whose connections are being used. In multi-tenant apps, set this per-request or per-user session.

CRUD Operations

All resources follow the same pattern: apideck.{api}.{resource}.{operation}().

// LIST - retrieve multiple records
const { data } = await apideck.crm.contacts.list({
  serviceId: "salesforce",
  limit: 20,
  filter: { email: "john@example.com" },
  sort: { by: "updated_at", direction: "desc" },
  fields: "id,name,email,phone_numbers",
});

// CREATE - create a new record
const { data: created } = await apideck.crm.contacts.create({
  serviceId: "salesforce",
  contact: {
    first_name: "John",
    last_name: "Doe",
    emails: [{ email: "john@example.com", type: "primary" }],
    phone_numbers: [{ number: "+1234567890", type: "mobile" }],
  },
});
console.log(created.id); // "contact_abc123"

// GET - retrieve a single record
const { data: contact } = await apideck.crm.contacts.get({
  id: "contact_abc123",
  serviceId: "salesforce",
});

// UPDATE - modify an existing record
const { data: updated } = await apideck.crm.contacts.update({
  id: "contact_abc123",
  serviceId: "salesforce",
  contact: { first_name: "Jane" },
});

// DELETE - remove a record
await apideck.crm.contacts.delete({
  id: "contact_abc123",
  serviceId: "salesforce",
});

Pagination

Use async iteration to automatically handle cursor-based pagination:

const result = await apideck.accounting.invoices.list({
  serviceId: "quickbooks",
  limit: 50,
});

// Automatically fetches next pages
for await (const page of result) {
  for (const invoice of page.data) {
    console.log(invoice.number, invoice.total);
  }
}

Or handle pagination manually:

let cursor: string | undefined;
do {
  const { data, meta } = await apideck.accounting.invoices.list({
    serviceId: "quickbooks",
    limit: 50,
    cursor,
  });
  for (const invoice of data) {
    console.log(invoice.number);
  }
  cursor = meta?.cursors?.next ?? undefined;
} while (cursor);

Error Handling

import { Apideck } from "@apideck/unify";
import * as errors from "@apideck/unify/models/errors";

try {
  const { data } = await apideck.crm.contacts.get({ id: "invalid" });
} catch (e) {
  if (e instanceof errors.BadRequestResponse) {
    console.error("Bad request:", e.message);
  } else if (e instanceof errors.UnauthorizedResponse) {
    console.error("Invalid API key or missing credentials");
  } else if (e instanceof errors.NotFoundResponse) {
    console.error("Record not found");
  } else if (e instanceof errors.PaymentRequiredResponse) {
    console.error("API limit reached or payment required");
  } else if (e instanceof errors.UnprocessableResponse) {
    console.error("Validation error:", e.message);
  } else {
    throw e;
  }
}

Common Parameters

Most list endpoints accept these parameters:

ParameterTypeDescription
serviceIdstringDownstream connector ID (e.g., "quickbooks", "salesforce")
limitnumberMax results per page (1-200, default 20)
cursorstringPagination cursor from previous response
filterobjectResource-specific filter criteria
sortobject`{by: string, direction: "asc" \"desc"}`
fieldsstringComma-separated field names to return
passThroughobjectPass-through query parameters for the downstream API

Pass-Through Parameters

When the unified model doesn't cover a connector-specific field, use passThrough:

const { data } = await apideck.accounting.invoices.list({
  serviceId: "quickbooks",
  passThrough: {
    search: "overdue",
  },
});

For creating/updating, use pass_through in the request body to send connector-specific fields:

const { data } = await apideck.crm.contacts.create({
  serviceId: "salesforce",
  contact: {
    first_name: "John",
    last_name: "Doe",
    pass_through: [
      {
        service_id: "salesforce",
        operation_id: "contactsAdd",
        extend_object: { custom_sf_field__c: "value" },
      },
    ],
  },
});

API Namespaces

The SDK organizes APIs by namespace. See the reference files for detailed endpoints:

NamespaceReferenceResources
apideck.accounting.*references/accounting-api.mdinvoices, bills, payments, customers, suppliers, ledgerAccounts, journalEntries, taxRates, creditNotes, purchaseOrders, balanceSheet, profitAndLoss, and more
apideck.crm.*references/crm-api.mdcontacts, companies, leads, opportunities, activities, notes, pipelines, users
apideck.hris.*references/hris-api.mdemployees, companies, departments, payrolls, timeOffRequests
apideck.fileStorage.*references/file-storage-api.mdfiles, folders, drives, driveGroups, sharedLinks, uploadSessions
apideck.ats.*references/ats-api.mdapplicants, applications, jobs
apideck.vault.*references/vault-api.mdconnections, connectionSettings, consumers, customMappings, logs, sessions
apideck.webhook.*references/webhook-api.mdwebhooks, eventLogs

Vault JS (Embeddable UI)

Use @apideck/vault-js to embed a pre-built modal that lets your users authorize connectors and manage integration settings. Session creation must happen server-side.

// 1. Server-side: create a session
const { data } = await apideck.vault.sessions.create({
  session: {
    consumer_metadata: { account_name: "Acme Corp", user_name: "John Doe", email: "john@acme.com" },
    redirect_uri: "https://myapp.com/integrations",
    settings: { unified_apis: ["accounting", "crm"] },
    theme: { vault_name: "My App", primary_color: "#4F46E5" },
  },
});

// 2. Client-side: open the modal
import { ApideckVault } from "@apideck/vault-js";

ApideckVault.open({
  token: sessionToken,
  onConnectionChange: (connection) => console.log("Changed:", connection),
  onClose: () => console.log("Closed"),
});

See references/vault-js.md for full configuration options, theming, React integration, and event callbacks.

平台分布

Codex

36.94%
按下载量换算39

Claude

32.96%
按下载量换算35

Cursor

18.04%
按下载量换算19

Gemini CLI

9.35%
按下载量换算10

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

接触敏感数据

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills