Agent skill

Cloudbase All In One

by TencentCloudBase in TencentCloudBase/CloudBase-AI-Toolkit

Unified CloudBase execution guide for all-in-one skill installs.

MITAuto-check passedDatabases

Install Cloudbase All In One

skills CLI
$ npx skills add TencentCloudBase/CloudBase-AI-Toolkit --skill cloudbase-all-in-one -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install TencentCloudBase/CloudBase-AI-Toolkit cloudbase-all-in-one --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/TencentCloudBase/CloudBase-AI-Toolkit.git skills-src && mkdir -p .claude/skills && cp -r skills-src/config/source/skills .claude/skills/cloudbase-all-in-one && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
cloudbase-all-in-one
GitHub stars
1.1k
Token cost
~3k tokens
SKILL.md length
1,325 words
Files
1
Skills in repo
49
Repo updated
First seen
Licence
MIT

At a glance

Unified CloudBase execution guide for all-in-one skill installs.

  • Works in 5 steps: BaaS-first, functions as last resort → Existing application with TODOs → Auth tasks → …
  • Tasks that involve NoSQL databases
  • SKILL.md covers Step 0 — Confirm the site…, Workflow, Scope and Activation Contract, plus 1 more section
  • Reaches tcb-api.tencentcloud.com and tcb-api.cloud.tencent.com

What it does

Cloudbase All In One is an agent skill from TencentCloudBase/CloudBase-AI-Toolkit. Unified CloudBase execution guide for all-in-one skill installs. Use this first for CloudBase app tasks, especially existing apps with TODOs, fixed pages, or active handlers. Routes PostgreSQL / CloudBase PG / app.rdb() / queryPgDatabase / managePgDatabase work away from legacy NoSQL and old auth patterns.

Its SKILL.md is about 3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Databases, covering NoSQL databases. It works with PostgreSQL and Model Context Protocol. The repository describes itself as: Backend for AI coding agents on CloudBase — database, auth, functions via Plugin, Skills & MCP. The licence is MIT.

When your agent uses it

  • Tasks that involve NoSQL databases

Example prompts

  • “/cloudbase-all-in-one”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. BaaS-first, functions as last resort
  2. Existing application with TODOs
  3. Auth tasks
  4. Database and storage tasks
  5. Targeted repair tasks

What it can do on your machine

Read from SKILL.md and the folder at commit ea2c202. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • tcb-api.tencentcloud.com
    • tcb-api.cloud.tencent.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Cloudbase All In One loads about 3k tokens when it runs. Until then it costs about 82 tokens; SKILL.md has 1,325 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~82
When it runs · the whole SKILL.md, loaded when a task matches
~3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from TencentCloudBase/CloudBase-AI-Toolkit at commit ea2c202, republished under its MIT licence (© TencentCloudBase). 1,325 words, ~3,012 tokens.

Download SKILL.mdSave it as .claude/skills/cloudbase-all-in-one/SKILL.md (or your agent's skills folder).
name
cloudbase-all-in-one
description
Unified CloudBase execution guide for all-in-one skill installs. Use this first for CloudBase app tasks, especially existing apps with TODOs, fixed pages, or active handlers. Routes PostgreSQL / CloudBase PG / app.rdb() / queryPgDatabase / managePgDatabase work away from legacy NoSQL and old auth patterns.
version
2.34.8
alwaysApply
true

CloudBase All-In-One

Step 0 — Confirm the site (domestic vs international)

CloudBase runs 国内站 (domestic, cloud.tencent.com) and 国际站 (international, tencentcloud.com) as two independent account systems — environments, consoles, API keys, and login state do not cross over. A wrong-site login looks like "logged in, but no environments" rather than a clear error, so settle the site before logging in or configuring MCP.

  • International (国际站) — connect the international remote MCP endpoint directly: https://tcb-api.tencentcloud.com/mcp/v1. For local stdio set TCB_SITE=intl + TCB_REGION=ap-singapore; for the tcb CLI set TCB_IS_INTL=true.
  • Domestic (国内站) — https://tcb-api.cloud.tencent.com/mcp/v1. Both switches are the defaults, so nothing extra to declare.
  • Two gaps to plan around: remote endpoints take no site / region query parameter (the hostname decides), and the international site has no NoSQL / document-database tools.

Workflow

Every CloudBase task follows this three-stage process:

1. Exploration  →  Read the matching skill completely before writing any code.
                   Search for it if needed, then Read the full SKILL.md content.
                   Only start implementing after understanding the API patterns,
                   pitfalls, and correct usage from the skill document.
2. Implementation
   ├── 2a. Resource preparation → Prefer MCP to prepare backend resources
   │     (enable auth providers, create database tables, configure storage domains,
   │      set up security rules — before writing frontend code).
   │     If MCP tools are missing in THIS session, configure MCP for next time and
   │     use `tcb` CLI now (`./cloudbase-cli/SKILL.md`).
   └── 2b. Frontend implementation → Write code, install deps, start server, test
3. Close-out  →  Run cloudbase-code-review, fix all errors, declare done

Key constraints:

  • Stage 2a (resource preparation) must come before frontend code. Prefer MCP tools; when they are not loaded yet (first session / pre-restart), fall back to tcb CLI after configuring MCP. Don't write frontend code that depends on uncreated resources (tables, auth providers, storage buckets) — the app cannot run end-to-end without them.
  • Do not skip Stage 3. The close-out catches known pitfalls that agents commonly miss during implementation.

Scope

Only handle tasks that are part of building, integrating, or maintaining a CloudBase application — including UI design, spec planning, and other development activities when they are in service of a CloudBase app. If the request has no CloudBase application context at all, stop and tell the user this skill only covers CloudBase-based development.

Activation Contract

Use this first when
  • The task is a CloudBase app build, integration, or repair and the workspace already contains an application implementation.
  • The request mixes auth, database, storage, and frontend work in one CloudBase application task.
Do this before broad exploration
  • Read the relevant skill before writing any code. Search for the matching skill with searchKnowledgeBase(mode="skill", skillName=...), then Read the returned file path to get the full SKILL.md content. Do not start implementing based only on the search result summary — the full document contains API signatures, parameter schemas, and pitfall warnings.
  • Inspect the existing implementation surfaces first:
    • src/lib/backend.*
    • src/lib/auth.*
    • src/lib/*service.*
    • route guards
    • the page handlers bound to the active form submit buttons
  • If these files contain TODOs, implement those TODOs in place before creating new helpers, examples, or replacement pages.
  • Do not start with UI redesign or design-spec output unless the user explicitly asks for visual changes.
  • Do not start with project-management loops such as repeated TaskCreate / TaskUpdate when the task is a single targeted repair. Read the active files and edit them directly.
Route quickly to the minimum needed skills
  • Minimal Web + database demo / Lovable-like BaaS fast path -> ./minimal-web-baas-demo/SKILL.md (default for 最小前后端 demo; BaaS-first, no cloud functions)
  • Web app execution -> ./web-development/SKILL.md
  • Declarative whole-project deploy from cloudbaserc (deployApply/deployPlan) -> ./cloudbase-declarative-deploy/SKILL.md
  • Web auth provider readiness -> ./auth-tool-cloudbase/SKILL.md
  • Web auth implementation -> ./auth-web-cloudbase/SKILL.md
  • CloudBase PostgreSQL / PG app data -> ./postgresql-development-cloudbase/SKILL.md
  • WeChat Pay / Official Account OAuth through CloudBase Integration Center -> ./cloudbase-wechat-integration/SKILL.md
  • Browser-side document database CRUD -> ./cloudbase-document-database-web-sdk/SKILL.md
  • Browser-side file upload -> ./cloud-storage-web/SKILL.md
  • Manage/operate underlying Tencent Cloud resources via cloud APIs (monitoring & alarms, CLB, CAM roles, cross-product infra) when no dedicated MCP tool exists -> ./cloud-api-operations/SKILL.md
  • Platform overview only when capability selection is still unclear -> ./cloudbase-platform/SKILL.md
  • If using searchKnowledgeBase(mode="skill"), pass the reference directory id such as postgresql-development-cloudbase or minimal-web-baas-demo, not a guessed alias.
High-yield guardrails
  • Prepare backend resources before writing frontend code. Prefer MCP for auth providers, database tables, storage domains, and security rules. If MCP tools are missing in this session, configure MCP for the next session and use tcb CLI now — do not stall waiting for restart. Frontend code written against non-existent resources will fail at runtime and require rework.
  • Change Safety Protocol: Before any non-trivial code or configuration change, you must strictly follow cloudbase-platform/references/protocols/change-safety-protocol.md (declare impact → obtain user confirmation → verify after change → escalate to root cause analysis after 3 occurrences of the same symptom).
  • Deployment Gate: Before any deployment, publish, custom domain, CloudRun, or public exposure work, you must complete the checks in cloudbase-platform/references/protocols/deployment-gate.md and present the mandatory declaration template.
  • If the same path fails 2-3 times, stop retrying and reroute. Check platform skill, auth domain, runtime, and permission model before editing more code.
  • Always specify EnvId explicitly in code, configuration, and command examples when initializing CloudBase clients or manager operations. Do not rely on the current CLI-selected environment or implicit defaults.
  • If the conversation only provides an environment alias, nickname, or other shorthand, resolve it with queryEnv(action=list, alias=..., aliasExact=true) and use the returned canonical full EnvId before calling auth.set_env, generating console links, or writing config/code. If the alias is ambiguous or missing, stop and ask the user to confirm.
Do NOT use this as
  • A reason to read every CloudBase skill before touching code.
  • A reason to start from platform overview when the existing code already reveals the stack and the missing pieces.
Show full SKILL.md (527 more words)Show less

Working rules

  1. BaaS-first, functions as last resort:

    • For 最小前后端 / Lovable-like demos, route first to ./minimal-web-baas-demo/SKILL.md. Order: connector ready → template warmup during credential wait → queryEnv → lock one DB → MCP schema → @cloudbase/js-sdk CRUD → preview. Default cloud function count = 0.
    • Before writing any cloud function or CloudRun service, ask: can the correct JS SDK surface handle this directly? Use db.collection(...).get() only for confirmed NoSQL collections; use app.rdb().from(...) for CloudBase PG tables; use auth / storage from the matching skill.
    • Use the matching JS SDK surface directly for: data reads/writes, file uploads, real-time updates, simple queries including leaderboards, lists, aggregations.
    • Only drop down to cloud functions when: the logic requires server-side permission enforcement that cannot be expressed in database rules/RLS, calling third-party services (payment, SMS, external APIs), or background jobs not triggered by the user.
    • Only drop down to CloudRun when: persistent connections (WebSocket), long-running compute, or custom runtimes are genuinely required.
  2. Existing application with TODOs:

    • Treat it as a targeted repair task, not a greenfield build.
    • Prefer the shortest path from current code to working flow.
  3. Auth tasks:

    • Read ./auth-web-cloudbase/SKILL.md before writing auth code — the full username/password recipe and provider readiness flow live there.
    • Blocking prerequisite for username-style accounts (admin, editor, any string without @): call queryAppAuth(action="getLoginConfig"); if loginMethods.usernamePassword !== true, enable it with manageAppAuth(action="patchLoginStrategy", patch={ usernamePassword: true }) first.
    • Use auth.signInWithPassword({ username, password }); never signUpWithEmailAndPassword / signInWithEmailAndPassword for username-style flows.
    • Once readiness is confirmed, return to the active frontend handler and finish the real login/register flow.
  4. Database and storage tasks:

    • Reuse the current shared app, auth, db, and storage helpers instead of creating parallel SDK wrappers.
    • If the task mentions CloudBase PG, PostgreSQL, Postgres, PG mode, JS SDK v3 PostgreSQL, app.rdb(), queryPgDatabase, managePgDatabase, mysqldb OpenAPI, or RLS, read ./postgresql-development-cloudbase/SKILL.md before touching database code.
    • When the PG task defines an access pattern or investigates a slow query, also read ./postgresql-best-practices-cloudbase/SKILL.md.
    • For CloudBase PG, use queryPgDatabase / managePgDatabase for schema and management; do not route PG work to MySQL queryMysqlDatabase / manageMysqlDatabase or NoSQL collection APIs.
    • For OpenAPI lookup, call searchKnowledgeBase({ mode: "openapi", apiName: "mysqldb" }) directly. Do not pass guessed action values such as getApiDocs or listEndpoints; those belong to no supported tool mode.
    • For CloudBase PG Web CRUD, prefer JS SDK v3 app.rdb() and documented storage app.storage.from() APIs before raw HTTP.
    • For browser-side CloudBase storage upload from local Vite/preview, check the actual browser host:port in security domains first (queryEnv(action="domains"), then envDomainManagement(action="create") if missing). A failed cover upload must not silently skip the subsequent PG article insert.
    • For CloudBase Web SDK db.collection(...).add(...), persist the created document ID from result._id.
    • For writes, validate the actual SDK result instead of assuming success.
    • Legacy API STOP card: If the task says PostgreSQL, CloudBase PG, PG mode, app.rdb(), queryPgDatabase, managePgDatabase, PostgREST, or RLS, do not write NoSQL examples from memory. Use app.rdb().from(...) for Web CRUD, queryPgDatabase / managePgDatabase for management, and auth.getSession() for Web auth guards. Do not use app.database(), db.collection(...), .where(), .orderBy(), app.uploadFile(), getLoginState(), or auth.getUser() as the PG/auth default.
  5. Targeted repair tasks:

    • Functional closure beats exploration.
    • Avoid broad repo sweeps, UI redesign, and detached demo code.
    • Keep file discovery narrow. Prefer direct reads of the known active files over Glob / broad search across the whole project.

© TencentCloudBase, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in config/source/skills of TencentCloudBase/CloudBase-AI-Toolkit.

Open the folder on GitHubat commit ea2c202

Compare with similar skills

Cloudbase All In One next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Cloudbase All In One compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Cloudbase All In One this skillTencentCloudBase/CloudBase-AI-Toolkit1.1k—~3kAutomated safety check: PassMIT
Mindsdb MCP SkillLeoYeAI/openclaw-master-skills2.2k—~2.5kAutomated safety check: PassMIT
Schema Explorationtimescale/pg-aiguide1.9k—~1.1kAutomated safety check: PassApache-2.0
Mongodb Query Optimizermongodb/agent-skills1902 repos~2.6kAutomated safety check: PassApache-2.0
Prisma Database Setupcurvenote/curvenote1693 repos~1.4kAutomated safety check: PassMIT
Neon Postgressmontlouis/bible-strong1711 repos~2.3kAutomated safety check: NotesGPL-3.0

Similar skills

  • Mindsdb MCP Skill

    LeoYeAI/openclaw-master-skills

    MindsDB MCP服务器交互技能,用于通过自然语言查询和操作200+企业级数据源。当用户需要查询数据库、分析数据、创建AI模型、连接数据源(MySQL、PostgreSQL、MongoDB、Excel、CSV、Gmail、Slack等)、执行SQL查询、进行数据预测、构建知识库(RAG)、智能问答、文档检索或任何与数据库交互的任务时使用此技能。即使没有明确提到MindsDB,只要涉及数据库操…

    2.2k GitHub stars~2.5k tokensUpdated 2 mo ago
    DatabasesAuto-check passed
  • Schema Exploration

    timescale/pg-aiguide

    Explore an existing PostgreSQL database before answering questions about its data or writing SQL.

    1.9k GitHub stars~1.1k tokensUpdated 6 days ago
    DatabasesAuto-check passed
  • Mongodb Query Optimizer

    mongodb/agent-skills

    Official

    Help with MongoDB query optimization and indexing. An agent skill from mongodb/agent-skills.

    190 GitHub starsUsed in 2 repos~2.6k tokens
    DatabasesAuto-check passed
  • Prisma Database Setup

    curvenote/curvenote

    Guides for configuring Prisma with different database providers (PostgreSQL, MySQL, SQLite, MongoDB, etc.).

    169 GitHub starsUsed in 3 repos~1.4k tokens
    DatabasesAuto-check passed
  • Neon Postgres

    smontlouis/bible-strong

    Guides and best practices for working with Lakebase Postgres, the database behind Neon.

    171 GitHub starsUsed in 1 repo~2.3k tokens
    DatabasesAuto-check: notes
  • A skill your agent uses to migrate identified PostgreSQL tables to Timescale/TimescaleDB hypertables with optimal configuration and validation.

    1.9k GitHub starsUsed in 1 repo~3.8k tokens
    DatabasesAuto-check: warnings

More from TencentCloudBase/CloudBase-AI-Toolkit

All 49 skills in this repo
  • AI Model Nodejs

    TencentCloudBase/CloudBase-AI-Toolkit

    A skill your agent uses for Node.js backend AI via @cloudbase/node-sdk (=3.16.0) — cloud functions, CloudRun, Express/Koa/NestJS, serverless APIs, scheduled jobs, LLM proxies, agent orchestration.

    1.1k GitHub starsUsed in 3 repos~5k tokens
    Auto-check passed
  • HTTP API Cloudbase

    TencentCloudBase/CloudBase-AI-Toolkit

    CloudBase official HTTP API client guide. An agent skill from TencentCloudBase/CloudBase-AI-Toolkit.

    1.1k GitHub starsUsed in 3 repos~2.1k tokens
    Auto-check passed
  • Cloud API Recipe Authoring

    TencentCloudBase/CloudBase-AI-Toolkit

    Author or revise a cloud-api-operations recipe (config/source/skills/cloud-api-operations/references/recipes/).

    1.1k GitHub stars~3.7k tokensUpdated yesterday
    Auto-check passed
  • Manage Local Skills

    TencentCloudBase/CloudBase-AI-Toolkit

    Analyze, standardize, validate, and sync locally maintained skills into agent skill directories with a skills CLI-aligned workflow.

    1.1k GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed
  • Cloudbase Agent Python

    TencentCloudBase/CloudBase-AI-Toolkit

    Build production-ready AI agent backends using the CloudBase Agent Python SDK — create agents with LangGraph/CrewAI/LlamaIndex, serve them via FastAPI with AG-UI protocol streaming +…

    1.1k GitHub starsUsed in 2 repos~2.9k tokens
    Auto-check: notes
  • Cloudbase

    TencentCloudBase/CloudBase-AI-Toolkit

    A skill your agent uses when you develop, design, build, deploy, debug, migrate, or troubleshoot CloudBase (腾讯云开发, 云开发, TCB, 微信云开发) projects — Web, 微信小程序, 小程序, uni-app, mobile (iOS, Android…

    1.1k GitHub starsUsed in 1 repo~4.7k tokens
    Auto-check passed

Categories

Questions about Cloudbase All In One

What does Cloudbase All In One do?

Unified CloudBase execution guide for all-in-one skill installs. Cloudbase All In One is an agent skill from TencentCloudBase/CloudBase-AI-Toolkit. Unified CloudBase execution guide for all-in-one skill installs.

When should I use Cloudbase All In One?

Cloudbase All In One fits situations like: tasks that involve NoSQL databases.

How do I install Cloudbase All In One in Claude Code?

Run `npx skills add TencentCloudBase/CloudBase-AI-Toolkit --skill cloudbase-all-in-one -a claude-code`. Or copy the skill folder (config/source/skills in TencentCloudBase/CloudBase-AI-Toolkit) into .claude/skills/cloudbase-all-in-one in your project. Claude Code loads it when a task matches its description.

How do I install Cloudbase All In One in Codex?

Run `npx skills add TencentCloudBase/CloudBase-AI-Toolkit --skill cloudbase-all-in-one -a codex`. Or copy the skill folder (config/source/skills in TencentCloudBase/CloudBase-AI-Toolkit) into .agents/skills/cloudbase-all-in-one in your project. Codex loads it when a task matches its description.

Can I use Cloudbase All In One in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add TencentCloudBase/CloudBase-AI-Toolkit --skill cloudbase-all-in-one -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cloudbase-all-in-one, .gemini/skills/cloudbase-all-in-one, .github/skills/cloudbase-all-in-one and .opencode/skills/cloudbase-all-in-one in your project.

What does Cloudbase All In One need to run?

SKILL.md names no scripts, command-line tools or credentials: Cloudbase All In One is instructions for the agent only.

Does Cloudbase All In One access the network?

SKILL.md names 2 domains. In commands or code: tcb-api.tencentcloud.com and tcb-api.cloud.tencent.com; the agent is likely to contact these when it follows the instructions. This is read from the text; nothing was executed.

Is Cloudbase All In One safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Cloudbase All In One use?

Cloudbase All In One is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Cloudbase All In One use?

About 3k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Cloudbase All In One?

Skills that share tags, products or a category with Cloudbase All In One: Mindsdb MCP Skill (LeoYeAI/openclaw-master-skills, 2.2k stars), Schema Exploration (timescale/pg-aiguide, 1.9k stars), Mongodb Query Optimizer (mongodb/agent-skills, 190 stars) and Prisma Database Setup (curvenote/curvenote, 169 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Cloudbase All In One?

TencentCloudBase (a GitHub organization) maintains it in TencentCloudBase/CloudBase-AI-Toolkit, which has 1,132 GitHub stars. The repository holds 49 skills in this directory. The repository was last updated on October 6, 2026.

Source: TencentCloudBase/CloudBase-AI-Toolkit on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.