Token导航 LogoToken导航

Agent Skill · 开发

cloud(cloud 搜索)

cloud 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。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。

cloud 是开发类 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,第三方扫描有未通过项,按安装量在开发类的 21,746 个 Skill 里排第 12,597。

总安装 1,212周安装 50GitHub Stars 公开资料未说明下载 396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:cloud(cloud 搜索)
来源仓库:https://github.com/shiplightai/agent-skills
仓库路径:skills/cloud
安装命令:
npx skills add https://github.com/shiplightai/agent-skills --skill cloud
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/shiplightai/agent-skills --skill cloud

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

cloud 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。它可结合来源仓库、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。适用于开发类任务,通过 npx skills add 命令从指定 GitHub 仓库安装。cloud 属于开发类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/cloud

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/shiplightai/agent-skills --skill cloud;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

cloud 怎么安装?

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

cloud 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。

cloud 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。

cloud 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。

cloud 用的人多吗?

公开数据是累计安装 1,212 次,近一周安装 50 次,下载 396 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。

cloud 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。
  • 需要处理开发类任务、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。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
cloud

cloud 搜索 · 本页

1,212—接触敏感数据未明确未通过

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

Shiplight Cloud

Sync local YAML test cases, templates, and TypeScript functions with the Shiplight cloud using MCP tools. Primarily used to push local tests up to the cloud (and pull cloud tests back down) so they can run on a schedule, be shared with your team, and integrate with CI. Also manages test runs, environments, folders, suites, and accounts via the REST API.

Setup

Requires a Shiplight cloud subscription and a SHIPLIGHT_API_TOKEN. If cloud MCP tools (save_test_case, get_test_case, etc.) are not in the tool list, the token is missing.

Tell the user:

Cloud tools are not available. Get your API token from https://app.shiplight.ai/settings/api-tokens, set SHIPLIGHT_API_TOKEN in your project's .env file, then reconnect MCP (/mcp).

If the user provides a token, append it to the project's .env file (create if needed) and tell them: "Saved to <project>/.env — make sure .env is in your .gitignore. Reconnect MCP (/mcp) to activate cloud tools."

All REST API calls require:

Authorization: Bearer $SHIPLIGHT_API_TOKEN

Error Handling

ErrorAction
401 UnauthorizedToken is invalid or expired — ask user to check SHIPLIGHT_API_TOKEN in .env
403 ForbiddenInsufficient permissions — inform user
404 Not FoundResource not found — report to user
422 ValidationShow validation message to user
Tool not foundToken is missing — guide user through setup above

MCP Tools

These tools are available when SHIPLIGHT_API_TOKEN is set. Prefer file_path over passing content directly (saves tokens). Always use output_format: 'yaml' for get_test_case.

  • Upload: save_test_case, save_test_account, save_template, save_function
  • Download: get_test_case, get_template, get_function
  • Account: save_test_account — create/update test account with optional storage_state_path to upload local browser session to cloud

ID Tracking

After uploading, add the returned cloud ID to the local file so future saves update instead of creating duplicates:

ArtifactLocal fileID field
Test case*.test.yamltest_case_id: 123 (top-level YAML field)
Templatetemplates/*.yamltemplate_id: 45 (top-level YAML field)
Functionhelpers/*.ts@function_id 67 (JSDoc tag per export)

REST API

Base URL: https://api.shiplight.ai

Test Cases

List Test Cases
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-cases

Query parameters:

ParamTypeDescription
idsstringComma-separated test case IDs
folderIdnumberFilter by exact folder
folderIdRecursivenumberFilter by folder and all descendants
labelIdsstringComma-separated label IDs (OR logic — matches test cases with ANY of the labels)
createdBystringFilter by creator user ID
orderBystringOrder by field (default: "id")
orderasc \descOrder direction (default: "desc")
limitnumberMax results to return

Response: {data: [{id, title, test_flow, folder_id,...}], count: number}

Get Test Case
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-cases/123

Response: {id, title, test_flow, folder_id}

Delete Test Case (soft delete)

Marks the test case and its results as deleted (soft delete — records are retained but hidden from queries).

curl -X DELETE -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-cases/123

Response: {success: true, message: "Test case deleted"}

Move Test Cases to Folder

Batch-update the folder assignment for multiple test cases. Set folder_id to null to move to root.

curl -X PUT -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"test_ids": [101, 102], "folder_id": 5}' \
  https://api.shiplight.ai/v1/test-cases/batch-update-folder

Body: {test_ids: number[], folder_id: number | null}

Response: {success: true, data: [/* updated test cases */], message: "Successfully updated 2 test cases"}


Test Runs

List Test Runs
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  "https://api.shiplight.ai/v1/test-runs?limit=10"

Query parameters:

ParamTypeDescription
testPlanIdnumberFilter by test plan
triggerstringFilter by trigger ("API", "MANUAL")
resultstringFilter by result ("PASSED", "FAILED")
limitnumberMax results to return

Response: array of {id, status, result, trigger, start_time, end_time, duration, total_test_case_count, passed_test_case_count, failed_test_case_count}

Get Test Run Details

Note: This endpoint has no /v1/ prefix.

curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/run-results/456

Response:

{
  "testRun": { "id": 456, "status": "COMPLETED", "result": "PASSED" },
  "testCaseResults": [
    { "id": 789, "test_case_id": 123, "result": "PASSED", "duration": 45 }
  ]
}
Trigger Test Run

Run a test case, test suite, or a combination in the cloud.

By test case:

curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"trigger": "API"}' \
  https://api.shiplight.ai/v1/test-run/test-case/123

By test suite:

curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"trigger": "API"}' \
  https://api.shiplight.ai/v1/test-run/test-suite/1

Generic (multiple test cases, suites, and/or labels):

curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"trigger": "API", "test_case_ids": [101, 102], "test_suite_ids": [1]}' \
  https://api.shiplight.ai/v1/test-run

By labels (run all test cases with any of the specified labels):

curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"trigger": "API", "label_ids": [22, 18]}' \
  https://api.shiplight.ai/v1/test-run

Body (all trigger endpoints):

FieldTypeDescription
triggerstringRequired. Use "API"
test_case_idsnumber[]Generic endpoint only — test case IDs to run
test_suite_idsnumber[]Generic endpoint only — test suite IDs to run
label_idsnumber[]Generic endpoint only — label IDs; resolves to test cases with ANY of these labels (OR logic). Can be combined with test_case_ids and test_suite_ids
environment{id?: string}Override environment

Response (201): test run object with {id, status, result,...}

After triggering, poll GET /v1/test-runs?limit=1 or GET /run-results/<id> to check status.

Get Test Case Result

Note: This endpoint has no /v1/ prefix.

curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/test-case-results/789

Response: {id, test_case_id, test_run_id, result, status, duration, environment_name, environment_url, video, trace, report_s3_uri, report, error}

The video, trace, and report_s3_uri fields contain S3 URIs — use the Artifacts endpoint to download them.

The report field contains step-by-step execution details in report[0].resultJson — each step has description, status, message, duration, and artifact S3 URIs (screenshot_s3_path, messages_s3_path, etc.).


Environments

List Environments
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/environments

Response: array of {id, name, url}

Get Environment
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/environments/1

Response: {id, name, url}


Variables

Environment-scoped variables — the cloud equivalent of variables in playwright.config.ts. Use isSensitive: true for secrets (passwords, API keys) so they're masked in logs.

List Variables
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  "https://api.shiplight.ai/v1/variables?environmentId=1"

Response: array of {id, name, value, environment_id, is_sensitive}


Test Accounts

List Test Accounts
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  "https://api.shiplight.ai/v1/test-accounts?environmentId=1"

Query: environmentId (number, optional but recommended) — filter by environment.

Response: array of {id, name, username, environmentId, loginConfig}

Get Test Account
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-accounts/1

Response: {id, name, username, password, environmentId, loginConfig}


Folders

List All Folders
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-folders/all

Optional query: ?search=keyword

Response: array of {id, name, description, parentId, pathIds}

List Folders by Parent
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  "https://api.shiplight.ai/v1/test-folders?parentId=1"

Omit parentId entirely for root-level folders.

Response: array of {id, name, description, parentId}

Get Folder
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-folders/1

Response: {id, name, description, parentId, pathIds}

Create Folder
curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "Smoke Tests", "parentId": 1}' \
  https://api.shiplight.ai/v1/test-folders

Body: {name: string, description?: string, parentId?: number | null}

Response (201): {success: true, data: {id, name, description, parentId}}

Update Folder
curl -X PATCH -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "Regression Tests"}' \
  https://api.shiplight.ai/v1/test-folders/1

Body: {name?: string, description?: string}

Response: {success: true, data: {id, name, description}}

Delete Folder
curl -X DELETE -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-folders/1

Response: {success: true}

Move Folder

Move a folder to a different parent. Set parentId to null to move to root.

curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"parentId": 5}' \
  https://api.shiplight.ai/v1/test-folders/1/move

Body: {parentId: number | null}

Response: {success: true}


Test Suites

Test suites group test cases for organized execution. Use suites to manage which test cases run together.

List All Suites
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-suites

Response: array of {id, title, description, testCount, createdAt, updatedAt}

Get Suite

Returns suite metadata and its test cases. Always use ?fields= to select only the columns you need — without it, the endpoint returns full test case entities which is too expensive.

?fields= accepts a comma-separated list of columns: id, title, description, status, folder_id, test_type, created_at, updated_at, url

curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  "https://api.shiplight.ai/v1/test-suites/1?fields=id,title,status"

Response:

{
  "suite": { "id": 1, "title": "Smoke Tests", "description": "...", "testCount": 5 },
  "testCases": [
    { "id": 101, "title": "Login flow", "status": "Active" },
    { "id": 102, "title": "Search", "status": "Active" }
  ]
}
Create Suite
curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"title": "Smoke Tests", "description": "Core user flows"}' \
  https://api.shiplight.ai/v1/test-suites

Body: {title: string, description?: string}

Response (201): {id, title, description, createdAt, updatedAt}

Update Suite
curl -X PUT -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"title": "Regression Tests"}' \
  https://api.shiplight.ai/v1/test-suites/1

Body: {title?: string, description?: string}

Response: {success: true, message: "Test suite updated successfully"}

Delete Suite
curl -X DELETE -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/test-suites/1

Response: {success: true, message: "Test suite deleted"}

Add Test Cases to Suite
curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"testCaseIds": [101, 102, 103]}' \
  https://api.shiplight.ai/v1/test-suites/1/test-cases

Body: {testCaseIds: number[]}

Response: {success: true, message: "Test cases added to suite successfully"}

Remove Test Cases from Suite
curl -X DELETE -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"testCaseIds": [101]}' \
  https://api.shiplight.ai/v1/test-suites/1/test-cases

Body: {testCaseIds: number[]}

Response: {success: true, message: "Test cases removed from suite successfully"}


Labels

Labels let you tag test cases (e.g., daily-regression, pre-merge) and trigger runs by label instead of manually managing suites.

List Labels
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/labels

Response: {success: true, data: [{id, name, color, organizationId, createdAt, updatedAt}]}

Create Label
curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "daily-regression", "color": "#4CAF50"}' \
  https://api.shiplight.ai/v1/labels

Body: {name: string, color: string}

Response (201): {success: true, data: {id, name, color}}

Update Label
curl -X PUT -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "nightly-regression"}' \
  https://api.shiplight.ai/v1/labels/22

Body: {name?: string, color?: string}

Delete Label
curl -X DELETE -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/labels/22
Add Labels to Test Case
curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"labelIds": [22, 18]}' \
  https://api.shiplight.ai/v1/labels/test-case/123/add

Body: {labelIds: number[]}

Remove Labels from Test Case
curl -X POST -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"labelIds": [22]}' \
  https://api.shiplight.ai/v1/labels/test-case/123/remove

Body: {labelIds: number[]}

Get Labels for Test Case
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/labels/test-case/123

Response: {success: true, data: [{id, name, color}]}

Get Test Case IDs by Label
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/labels/test-cases/22

Response: {success: true, data: [101, 102, 103]}


Templates (Reusable Steps)

Reusable test step sequences that can be referenced from test cases via template: in YAML or reference_id in the cloud. The API uses the legacy name "reusable-steps".

Get Template
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/reusable-steps/138

Response: {id, name, description, statements}


Test Functions

Custom TypeScript functions that can be called from test cases via call: "file#export" in YAML.

Get Test Function
curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  https://api.shiplight.ai/v1/functions/42

Response: {id, name, description, code, status}


Artifacts

Download S3 File

Download test artifacts (videos, traces, reports) referenced by S3 URIs in test case results.

curl -H "Authorization: Bearer $SHIPLIGHT_API_TOKEN" \
  "https://api.shiplight.ai/v1/s3/file?uri=s3://bucket/path/video.webm"

Query: uri (string, required) — S3 URI from test result fields (video, trace, report_s3_uri).

Response: raw file contents with appropriate content-type. For binary files (videos, traces), use curl -o <output_file> to save to disk.


Workflows

Sync local project to cloud

  1. Sync test accounts: call save_test_account with credentials and storage_state_path (if the local project uses storageState-based auth)
  2. Sync test cases: find all .test.yaml files, call save_test_case with file_path for each. If the file already has a test_case_id, it updates; if not, it creates and returns an ID.
  3. Track IDs: add the returned test_case_id to the YAML if not already present
  4. Sync templates and functions: repeat with save_template and save_function

Download test case from cloud

  1. Call get_test_case with test_case_id and output_format: 'yaml'
  2. Convert the cloud url: to baseURL + relative paths. Cloud test cases have a top-level url: field for the starting URL (e.g., url: https://app.example.com/dashboard). Local YAML tests split this into two parts: Remove the top-level url: field from the downloaded YAML, ensure baseURL is configured, and convert any absolute URL statements to relative paths.

- Base URL — set once in playwright.config.ts at the project level (use: {baseURL: 'https://app.example.com'}) if all tests share the same origin, or per-test via base_url: https://app.example.com. Prefer project-level. - Navigation — use relative paths in the test steps (URL: /dashboard).

  1. Save the returned YAML to a .test.yaml file

Download test artifacts

  1. GET /test-case-results/<id> → find video, trace, or report_s3_uri fields (S3 URIs)
  2. GET /v1/s3/file?uri=<S3_URI> → download the artifact content

平台分布

Codex

37.6%
按下载量换算149

Claude

29.88%
按下载量换算118

Cursor

17.8%
按下载量换算70

Gemini CLI

8.91%
按下载量换算35

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

未通过

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