CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cloudbase

Use this skill 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, Flutter, React Native). Covers UI (页面, 界面, 表单, dashboard, prototype, 原型); auth (登录, 注册, OAuth, publishable key); databases (NoSQL 文档数据库, PostgreSQL/CloudBase PG, 已有 MySQL, app.rdb(), queryPgDatabase/managePgDatabase, CRUD, security rules); 云函数/cloud functions (serverless, scf_bootstrap); CloudRun (云托管, Dockerfile); 云存储; built-in AI (内置大模型, AI 对话, streaming, 流式输出, 图片生成, generateText, streamText, createModel, generateImage, TokenHub, Hunyuan, DeepSeek, GLM, Kimi, Token Credits 资源包, 小程序成长计划); third-party/custom model onboarding (第三方大模型接入, 大模型调用, LLM API); AI agent (智能体, AG-UI, LangGraph); ops troubleshooting (巡检, 诊断, 日志); spec workflow (需求文档, 技术方案, requirements, tasks.md). Do NOT use for non-CloudBase projects, pure frontend without CloudBase, or self-hosted backends without CloudBase.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Failed to scan

The risk profile of this skill

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-engineered entry/routing file: exact commands and tool invocations, an explicit ordered workflow with mandatory validation and reroute rules, and disciplined deferral of scenario detail to child skills. The main gaps are redundant triple coverage of reference navigation and minor reference-index inconsistencies.

Suggestions

Consolidate the 'Reference Files Location' tree, 'On-demand references', and 'Reference index' sections into a single annotated reference list to remove redundant navigation content.

Add references/international-site.md to the 'Reference index' (it is referenced in the body and exists in the bundle), and consider giving routing-table skill ids a consistent path-resolvable form.

DimensionReasoningScore

Conciseness

Quotes: "Always specify `EnvId` explicitly; do not rely on CLI-selected or implicit env state", "Load only when needed (do not expand this entry)". The body is dense and directive with almost no explanation of concepts Claude already knows, but reference navigation is covered three separate times ("📁 Reference Files Location" tree, "On-demand references", "Reference index"), which is redundant. Not 5 because of that triple-listed padding; not 3 because the rest of the content earns its tokens.

4 / 5

Actionability

Quotes: "`npx plugins add TencentCloudBase/cloudbase-plugin -y --scope user`", "`TCB_SITE=intl` + `TCB_REGION=ap-singapore`", "`queryEnv(action=list, alias=..., aliasExact=true)`", "`manageApps(action=\"deployApp\", ...)`", "`JSON.stringify(result, null, 2)`". Copy-paste-ready commands and exact tool invocations cover the entry file's own responsibilities, and the routing table gives exact skill ids with per-scenario pre-action checks. Not 4: the concrete commands are complete and cover the common cases for this entry/routing file.

5 / 5

Workflow Clarity

Quotes: "1. Exploration → ... 2a. Resource preparation → ... 2b. Frontend implementation → 3. Close-out", "Stage 2a must precede frontend code. Stage 3 is mandatory.", "After 2–3 failed attempts on the same path, stop and reroute", "Self-verify before claiming done. Static (`tsc` / lint / build / tests) and runtime (`agent-browser` ...)". Clear sequence with explicit validation steps and an error-recovery feedback loop. Not 4: validation checkpoints are explicit, not implicit.

5 / 5

Progressive Disclosure

Quotes: "Load only when needed (do not expand this entry)", "Keep scenario-specific pitfalls in child skills — do not expand this entry file", "Routing uses stable skill ids". Strong disclosure discipline, and all referenced paths (references/mcp-setup.md, references/site-onboarding.md, references/tooling-fallback.md, references/international-site.md, references/activation-map.yaml, references/console-links.md, references/deployment-workflow.md, references/scenarios.md, references/auth-web-cloudbase/SKILL.md, references/cloudbase-cli/SKILL.md) verified present in the bundle. Not 5: the routing table uses bare skill ids that must be resolved via the Activation Contract rule, reference navigation is spread across three overlapping sections, and references/international-site.md is referenced in the body but missing from the "Reference index" list.

4 / 5

Total

18

/

20

Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A highly specific, bilingual, trigger-rich description that explicitly answers both what and when, with a clear negative boundary. The only deduction is the second-person voice ("when you develop"), which the rubric penalizes on specificity.

DimensionReasoningScore

Specificity

Quotes: "develop, design, build, deploy, debug, migrate, or troubleshoot CloudBase ... projects", "app.rdb(), queryPgDatabase/managePgDatabase", "generateText, streamText, createModel, generateImage". The breadth of concrete actions and named APIs matches the anchor-5 comprehensive example, but the second-person phrasing "Use this skill when you develop..." triggers the rubric's 1-point voice penalty on specificity. Not 3: it lists far more than 1-2 concrete actions with comprehensive capability coverage.

4 / 5

Completeness

Quotes: "Use this skill when you develop, design, build, deploy, debug, migrate, or troubleshoot CloudBase ... projects" (when), "Covers UI ...; auth ...; databases ..." (what), and "Do NOT use for non-CloudBase projects, pure frontend without CloudBase, or self-hosted backends without CloudBase" (explicit negative boundary). Both what and when are explicit with concrete trigger phrases. Not 4: neither half could be more explicit.

5 / 5

Trigger Term Quality

Quotes: "腾讯云开发, 云开发, TCB, 微信云开发", "云函数/cloud functions (serverless, scf_bootstrap)", "Hunyuan, DeepSeek, GLM, Kimi", "巡检, 诊断, 日志". Comprehensive coverage of natural terms in both Chinese and English, including synonyms, product names, and API names users would actually say. Not 4: very few natural variants are missing.

5 / 5

Distinctiveness Conflict Risk

Quotes: "CloudBase (腾讯云开发, 云开发, TCB, 微信云开发) projects", "Do NOT use for non-CloudBase projects, pure frontend without CloudBase, or self-hosted backends without CloudBase". Clear niche with brand-specific triggers and an explicit exclusion clause; minimal conflict risk with other skills. Not 4: the exclusions remove even minor overlap risk.

5 / 5

Total

19

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

referenced_paths_exist

Referenced path issues: 2 deeper-than-1-level

Warning

Total

14

/

16

Passed

Repository

TencentCloudBase/CloudBase-AI-Toolkit

Path

connectors/cloudbase-intl/skills/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.