Token导航 LogoToken导航

Agent Skill · 研究检索

datocms-cda(数据管理系统)

datocms-cda 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。

datocms-cda 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 36,169。

总安装 1,972周安装 79GitHub Stars 公开资料未说明下载 638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:datocms-cda(数据管理系统)
来源仓库:https://github.com/datocms/agent-skills
仓库路径:skills/datocms-cda
安装命令:
npx skills add https://github.com/datocms/agent-skills --skill datocms-cda
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/datocms/agent-skills --skill datocms-cda

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。

用于查询 DatoCMS 内容交付 API (CDA) 的只读 GraphQL 接口。自动识别项目上下文并建立安全连接,无需手动配置 token。支持内容模型发现、记录查询和字段类型推断。datocms-cda 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

严格遵循步骤顺序,确保查询语句符合 CDA 规范要求。注意:CDA 为只读 API,内容修改必须通过 CMA 完成。

来源仓库内路径:skills/datocms-cda

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/datocms/agent-skills --skill datocms-cda;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

datocms-cda 怎么安装?

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

datocms-cda 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。

datocms-cda 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。

datocms-cda 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。

datocms-cda 用的人多吗?

公开数据是累计安装 1,972 次,近一周安装 79 次,下载 638 次。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。

datocms-cda 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
datocms-cda

数据管理系统 · 本页

1,972—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

DatoCMS Content Delivery API Skill

You are an expert at querying the DatoCMS Content Delivery API (CDA) using @datocms/cda-client. The CDA is a read-only GraphQL API — it has no mutations. All content changes go through the CMA (Content Management API). Follow these steps in order. Do not skip steps.


Step 1: Detect Context

If the project context is already established in this conversation (client package, token variable, framework, type generation setup), skip broad detection below. Re-inspect only when a question cannot be answered from prior context.

Silently examine the project to determine setup and configuration.

  1. Read package.json and check for @datocms/cda-client.

- If not installed, recommend: npm install @datocms/cda-client

  1. Search for existing executeQuery or rawExecuteQuery imports to understand how the project already uses the CDA client.
  2. Check .env, .env.local, or similar files for a DatoCMS API token. Look for variable names like:

- DATOCMS_CDA_TOKEN - DATOCMS_READONLY_TOKEN - DATOCMS_API_TOKEN - NEXT_PUBLIC_DATOCMS_CDA_TOKEN

  1. Check the framework context (Next.js, Astro, Remix, Nuxt, SvelteKit, etc.) to determine whether queries run on the server or client. CDA queries work in both environments, but tokens should not be exposed to the browser unless the project intentionally uses a public read-only token.
  2. Check for existing type generation setup:

- gql.tada: Look for gql.tada in package.json dependencies and an initGraphQLTada call (typically in lib/datocms/graphql.ts) - graphql-codegen: Look for @graphql-codegen/cli in devDependencies and a graphql.config.ts file - This detection is for context only — use it to write queries that match the project's existing setup (e.g., using the project's graphql() function instead of plain strings). Do not proactively suggest setting up type generation.

Important: The CDA needs a read-only API token (or a full-access CMA token, which also works). If you see a token named DATOCMS_API_TOKEN used for CMA operations, the user may need a separate read-only token for the CDA, or they can reuse the CMA token if appropriate.


Step 2: Understand the Task

Classify the user's task into one or more categories:

CategoryExamples
Basic queryingFetch records by slug/ID, query single-instance models, list collections
FilteringFilter by field values, AND/OR logic, meta field filters, deep filtering
Pagination & orderingPaginate large collections, sort results, tree/hierarchical queries
LocalizationQuery localized fields, fallback locales, all-locale values
Modular contentQuery block fields with GraphQL fragments, nested blocks
Structured textQuery DAST value/blocks/links, render with framework components
Images & mediaResponsive images, imgix transforms, placeholders, focal points, video
SEO & meta_seoMetaTags, favicons, globalSeo, Open Graph tags
Draft/preview & cachingDraft mode, strict mode, cache tags, CDN invalidation, Content Link
Type generationSet up gql.tada, configure graphql-codegen, generate schema types, typed queries

If the user's request is clear, skip clarifying questions and proceed directly.


Step 3: Load References

Based on the task classification, read the appropriate reference files from the references/ directory next to this skill file. Always load the core client reference. Only load what is relevant — do not load everything.

Always load:

  • references/client-and-config.md — Client setup, options, error handling, limits, scalars

Load per category:

Task categoryReference file
Basic querying (records, collections, meta)references/querying-basics.md
Filtering (field filters, AND/OR, deep filtering, uploads)references/filtering.md
Pagination & ordering (first/skip, auto-pagination, trees)references/pagination-and-ordering.md
Localizationreferences/localization.md
Modular content (blocks, fragments)references/modular-content.md
Structured text (DAST, rendering)references/structured-text.md
Images & media (responsiveImage, video)references/images-and-videos.md
SEO & meta tagsreferences/seo-and-meta.md
Draft/preview, caching, environments, Content Linkreferences/draft-caching-environments.md
Type generation (gql.tada, graphql-codegen, schema types)references/type-generation.md

Load cross-cutting references when needed:

  • If filtering localized fields → also load references/localization.md
  • If querying modular content inside structured text → also load references/modular-content.md
  • If querying images inside blocks → also load references/images-and-videos.md
  • If paginating a large filtered collection → also load references/pagination-and-ordering.md
  • If the query involves complex nesting → also load references/pagination-and-ordering.md for complexity costs

Step 4: Generate Code

Write the code following these mandatory rules:

Client Usage

  • Default to executeQuery from @datocms/cda-client, or the repo's existing wrapper around it (not raw fetch)
  • Use buildRequestHeaders() / buildRequestInit() when the framework needs integrated fetch handling, request tagging, or custom request plumbing
  • Use executeQueryWithAutoPagination when fetching more than 500 records
  • Use rawExecuteQuery only when you need response headers (e.g., cache tags)
  • Store the API token in an environment variable — never hardcode it

GraphQL Queries

  • Write queries as template literal strings (unless the project uses TypedDocumentNode / gql.tada)
  • Use GraphQL variables for all dynamic values — never use string interpolation in queries
  • Request only the fields you need — do not over-fetch
  • Use DatoCMS custom scalars in variable declarations (e.g., $first: IntType, $id: ItemId)

Structured Text

  • Always query all relevant sub-fields (value, blocks, links, inlineBlocks) when the structured text field uses them — omitting any causes silent data loss

Error Handling

  • Catch ApiError from @datocms/cda-client at appropriate boundaries
  • Do not add custom retry logic — autoRetry handles rate limits automatically

TypeScript

  • Follow the TypeScript strictness rules: no as unknown as, no unnecessary as casts
  • Let TypeScript infer types wherever possible
  • Use import type {...} for type-only imports

Step 5: Verify

Before presenting the final code:

  1. Token — Ensure the token comes from an environment variable and has read permissions
  2. Error handling — Ensure ApiError is caught at appropriate boundaries
  3. Pagination — If the collection could exceed 500 records, use executeQueryWithAutoPagination
  4. Draft mode — If includeDrafts is used, ensure it is intentional (not accidentally showing unpublished content in production)
  5. excludeInvalid — Recommend for stable schemas. If the schema is changing (migrations, new required fields), use filter: {_isValid: {eq: true}} instead to avoid re-validation errors
  6. Type safety — No type assertions (as) used to silence errors
  7. Imports — CDA client imports come from @datocms/cda-client; keep project-generated GraphQL helper imports when type generation is already wired
  8. Variables — All dynamic values use GraphQL variables, not string interpolation
  9. Structured text completeness — If querying structured text, all relevant sub-fields (value, blocks, links, inlineBlocks) are included
  10. Fetch integration — If the solution uses framework-native fetch, ensure CDA headers/init come from buildRequestHeaders() / buildRequestInit() instead of hand-rolled request wiring
  11. Type generation — If the project uses gql.tada or graphql-codegen, ensure queries use the project's graphql() function (not plain template literal strings) and that scalar mappings are configured

Cross-Skill Routing

This skill covers reading content via the GraphQL CDA. If the task involves any of the following, activate the companion skill:

ConditionRoute to
Mutating content, managing schema/uploads/webhooks, writing scripts (including querying records via REST for scripts)datocms-cma
Setting up draft mode endpoints, Web Previews, Content Link, real-time subscriptions, or cache tags in a frameworkdatocms-frontend-integrations
Building a DatoCMS plugindatocms-plugin-builder

平台分布

Codex

35.42%
按下载量换算226

Claude

31.24%
按下载量换算199

Cursor

21.77%
按下载量换算139

Gemini CLI

9.42%
按下载量换算60

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