When to Use
User needs help choosing, comparing, implementing, or de-risking something in the Tencent ecosystem. Use this for Tencent Cloud architecture, WeChat and WeCom decisions, Mini Program planning, payments and ads context, China-versus-global rollouts, and official-doc verification when Tencent naming is ambiguous.
Do not use it as a general China market strategy skill or as a substitute for dedicated product skills when the task is already narrowed to a single tool.
Architecture
Memory lives in ~/tencent/. If ~/tencent/ does not exist, run setup.md. See memory-template.md for structure.
CODEBLOCK0
Quick Reference
Use the smallest relevant file for the current task.
| Topic | File | Use it for |
|---|
| Setup flow and saved defaults | INLINECODE4 | Initialize local state and activation rules |
| Memory schema and status values |
memory-template.md | Create baseline local files |
| Tencent surface-area map |
ecosystem-map.md | Route requests to the right Tencent product family |
| Product and vendor choice |
decision-framework.md | Turn vague Tencent requests into a decision path |
| Tencent Cloud planning |
cloud-platform.md | Map workloads, regions, and operations to Tencent Cloud |
| Mainland-versus-global constraints |
mainland-vs-global.md | Check region, residency, language, and go-live risks |
| Official-source handling |
source-validation.md | Verify docs, translations, and time-sensitive claims |
| Delivery and escalation checklist |
rollout-checklist.md | Final review before recommending implementation or launch |
Requirements
- - No credentials required for planning, research, or source verification
- Account-specific execution may require user-approved access to Tencent Cloud, WeChat, or partner consoles
- Never ask the user to paste passwords, QR-login secrets, SMS codes, cookies, or private access keys into chat
Data Storage
Save only durable context that improves later Tencent work:
- - product families the user actually uses
- account or tenant names that the user explicitly shares
- preferred regions, languages, and documentation sources
- repeated compliance or rollout constraints
- final decisions with open risks and follow-up items
Do not store secrets, billing exports, raw customer data, or copied console tokens.
Core Rules
1. Lock the Tencent Surface First
- - Start by classifying the request into one primary surface: Tencent corporate, Tencent Cloud, WeChat ecosystem, business collaboration, ads and growth, payments, or games and content.
- If the user mixes several surfaces, split the answer into separate workstreams instead of pretending one Tencent path covers all of them.
- Most bad recommendations happen because "Tencent" was left vague.
2. Make Region and Audience Explicit
- - Every Tencent recommendation must state whether it targets mainland China, Hong Kong, Singapore, or broader international users.
- Also state the audience: consumers, enterprise teams, developers, advertisers, merchants, or operations.
- Product availability, docs, onboarding, payments, and compliance differ sharply by region and audience.
3. Prefer Official Sources, Then Reconcile Translations
- - For product behavior, limits, pricing, regulatory statements, SDK support, or launch requirements, start from Tencent-owned documentation and product pages.
- If English and Chinese sources diverge, record the mismatch and prefer the fresher or more specific original source.
- Never present translated summaries as stronger evidence than the original product documentation.
4. Translate Business Goals Into Product Decisions
- - Convert vague asks like "We need Tencent for China" into a concrete job: messaging, identity, payments, compute, object storage, ads, distribution, or local operations.
- Use the matrix in
decision-framework.md before naming tools. - Do not dump product lists when the user really needs a decision.
5. Treat Tencent Cloud as Its Own Platform
- - Map workloads using Tencent Cloud services and operational realities, not AWS names pasted onto a different console.
- Call out region support, account boundaries, observability gaps, and managed-service tradeoffs before recommending architecture.
- If a workload depends on mainland data residency, ICP, or cross-border traffic, say so up front.
6. Separate Research Guidance From Execution Rights
- - Planning and verification are safe defaults.
- Any step that would log in, change cloud resources, submit app assets, or touch payments must wait for explicit user approval and the correct account context.
- Never assume the same operator owns Tencent Cloud, WeChat OA, WeCom, ads, and payments.
7. End With a Decision Record
- - Finish every non-trivial task with a concise decision record: chosen path, rejected paths, assumptions, hard blockers, and what still needs human confirmation.
- Save only that durable decision state under
~/tencent/. - This keeps future Tencent work consistent instead of repeating the same ambiguity.
Common Traps
- - Treating Tencent, Tencent Cloud, WeChat, WeCom, and QQ as one interchangeable product family -> wrong routing and wrong owners
- Assuming international docs are complete mirrors of mainland documentation -> missing limits, onboarding steps, or pricing details
- Recommending Tencent Cloud by AWS analogy alone -> service names may look familiar while operations differ materially
- Ignoring mainland-versus-global rollout constraints until the end -> late blockers around residency, ICP, language, or payments
- Using unofficial English blog posts as primary evidence -> outdated or simplified guidance becomes product truth
- Assuming one enterprise account can approve all Tencent surfaces -> ownership and permissions are usually fragmented
External Endpoints
| Endpoint | Data Sent | Purpose |
|---|
| https://www.tencent.com | Query text and page requests | Corporate and product-overview verification |
| https://cloud.tencent.com |
Query text and page requests | Mainland Tencent Cloud documentation and pricing |
| https://www.tencentcloud.com | Query text and page requests | International Tencent Cloud documentation and product availability |
| https://developers.weixin.qq.com | Query text and page requests | WeChat Mini Program and ecosystem developer documentation |
| https://work.weixin.qq.com | Query text and page requests | WeCom product and admin documentation |
No other endpoints should be used unless the user explicitly approves additional sources or account-specific execution.
Security & Privacy
Data that may leave your machine:
- - search terms and page requests sent to Tencent-owned documentation or product sites
- optional notes sent to user-approved comparison sources when official docs are incomplete
Data that stays local:
- - activation preferences and Tencent work history under INLINECODE14
- account labels, region defaults, and decision records saved in local markdown files
- trusted-source notes and rollout constraints
This skill does NOT:
- - store secrets in local markdown files
- claim a product is available everywhere without checking region
- log into Tencent surfaces without explicit user approval
- rewrite its own skill definition files
Trust
This skill depends on Tencent-owned sites and any approved supporting sources used for verification.
Only install and use it if you trust those services with your research queries and planning workflow.
Scope
This skill ONLY:
- - routes Tencent-related requests to the correct product family
- compares Tencent options using official-source verification
- plans Tencent Cloud, WeChat ecosystem, and region-sensitive rollouts
- keeps lightweight local memory in INLINECODE15
This skill NEVER:
- - invent product capabilities or region availability
- treat unofficial summaries as final evidence
- request passwords, QR login tokens, cookies, or raw secrets
- execute account-changing actions without explicit user approval
Related Skills
Install with
clawhub install <slug> if user confirms:
- -
cloud - General cloud architecture and platform tradeoff guidance - INLINECODE18 - Object storage design, migration, and operations patterns
- INLINECODE19 - Payment flow design, merchant risk checks, and launch discipline
- INLINECODE20 - Broader market and source-comparison workflows
- INLINECODE21 - Competitive and market-level framing beyond product routing
Feedback
- - If useful: INLINECODE22
- Stay updated: INLINECODE23
何时使用
用户需要在腾讯生态系统中选择、比较、实施或降低风险时使用此技能。适用于腾讯云架构、微信和企业微信决策、小程序规划、支付和广告场景、中国与全球部署,以及在腾讯命名不明确时进行官方文档验证。
当任务已缩小到单一工具时,请勿将其用作通用中国市场策略技能或专用产品技能的替代品。
架构
记忆存储在 ~/tencent/ 目录中。如果 ~/tencent/ 不存在,请运行 setup.md。结构请参见 memory-template.md。
text
~/tencent/
|- memory.md # 激活规则、产品范围和首选输出
|- accounts.md # 已知租户、产品和所有权边界
|- regions.md # 中国大陆、香港、新加坡和全球部署说明
|- sources.md # 官方文档、可信翻译和弱来源警告
- - decisions.md # 最终建议、假设和未决风险
快速参考
为当前任务使用最小的相关文件。
| 主题 | 文件 | 用途 |
|---|
| 设置流程和已保存的默认值 | setup.md | 初始化本地状态和激活规则 |
| 记忆模式和状态值 |
memory-template.md | 创建基线本地文件 |
| 腾讯覆盖面地图 | ecosystem-map.md | 将请求路由到正确的腾讯产品系列 |
| 产品和供应商选择 | decision-framework.md | 将模糊的腾讯请求转化为决策路径 |
| 腾讯云规划 | cloud-platform.md | 将工作负载、区域和运营映射到腾讯云 |
| 中国大陆与全球约束 | mainland-vs-global.md | 检查区域、居住地、语言和上线风险 |
| 官方来源处理 | source-validation.md | 验证文档、翻译和时效性声明 |
| 交付和升级清单 | rollout-checklist.md | 在推荐实施或发布前进行最终审查 |
要求
- - 规划、研究或来源验证无需凭证
- 特定账户的执行可能需要用户批准的腾讯云、微信或合作伙伴控制台访问权限
- 切勿要求用户将密码、二维码登录密钥、短信验证码、Cookie 或私有访问密钥粘贴到聊天中
数据存储
仅保存可改善后续腾讯工作的持久上下文:
- - 用户实际使用的产品系列
- 用户明确分享的账户或租户名称
- 首选区域、语言和文档来源
- 重复出现的合规或部署约束
- 包含未决风险和后续事项的最终决策
不存储密钥、账单导出、原始客户数据或复制的控制台令牌。
核心规则
1. 首先锁定腾讯覆盖面
- - 首先将请求归类到一个主要覆盖面:腾讯公司、腾讯云、微信生态系统、企业协作、广告与增长、支付、或游戏与内容。
- 如果用户混合了多个覆盖面,将答案拆分为独立的工作流,而不是假装一个腾讯路径可以覆盖所有。
- 大多数糟糕的建议都源于腾讯一词表述模糊。
2. 明确区域和受众
- - 每项腾讯建议必须说明其目标是中国大陆、香港、新加坡还是更广泛的国际用户。
- 同时说明受众:消费者、企业团队、开发者、广告主、商家或运营人员。
- 产品可用性、文档、接入、支付和合规性因区域和受众而有显著差异。
3. 优先使用官方来源,然后协调翻译
- - 对于产品行为、限制、定价、监管声明、SDK 支持或发布要求,从腾讯自有文档和产品页面入手。
- 如果英文和中文来源存在差异,记录不一致之处,并优先采用更新或更具体的原始来源。
- 切勿将翻译摘要呈现为比原始产品文档更有力的证据。
4. 将业务目标转化为产品决策
- - 将我们需要腾讯进入中国等模糊请求转化为具体任务:消息、身份、支付、计算、对象存储、广告、分发或本地运营。
- 在命名工具之前,使用 decision-framework.md 中的矩阵。
- 当用户真正需要决策时,不要罗列产品列表。
5. 将腾讯云视为独立平台
- - 使用腾讯云服务和运营现实来映射工作负载,而不是将 AWS 名称粘贴到不同的控制台上。
- 在推荐架构之前,明确指出区域支持、账户边界、可观测性差距和托管服务权衡。
- 如果工作负载依赖于中国大陆数据驻留、ICP 或跨境流量,请提前说明。
6. 区分研究指导与执行权限
- - 规划和验证是安全的默认操作。
- 任何涉及登录、更改云资源、提交应用资产或接触支付的操作,都必须等待用户明确批准和正确的账户上下文。
- 切勿假设同一操作者拥有腾讯云、微信公众号、企业微信、广告和支付的所有权限。
7. 以决策记录结束
- - 每项重要任务都以简洁的决策记录结束:所选路径、被拒绝的路径、假设、硬性障碍以及仍需人工确认的事项。
- 仅将该持久决策状态保存在 ~/tencent/ 下。
- 这使未来的腾讯工作保持一致,而不是重复相同的模糊性。
常见陷阱
- - 将腾讯、腾讯云、微信、企业微信和 QQ 视为一个可互换的产品系列 -> 错误的路由和错误的所有者
- 假设国际文档是中国大陆文档的完整镜像 -> 遗漏限制、接入步骤或定价细节
- 仅通过 AWS 类比推荐腾讯云 -> 服务名称可能看起来很熟悉,但运营方式存在实质性差异
- 直到最后才考虑中国大陆与全球部署的约束 -> 关于居住地、ICP、语言或支付的后期障碍
- 使用非官方英文博客文章作为主要证据 -> 过时或简化的指南成为产品事实
- 假设一个企业账户可以批准所有腾讯覆盖面 -> 所有权和权限通常是分散的
外部端点
| 端点 | 发送的数据 | 用途 |
|---|
| https://www.tencent.com | 查询文本和页面请求 | 公司和产品概览验证 |
| https://cloud.tencent.com |
查询文本和页面请求 | 中国大陆腾讯云文档和定价 |
| https://www.tencentcloud.com | 查询文本和页面请求 | 国际腾讯云文档和产品可用性 |
| https://developers.weixin.qq.com | 查询文本和页面请求 | 微信小程序和生态系统开发者文档 |
| https://work.weixin.qq.com | 查询文本和页面请求 | 企业微信产品和管理员文档 |
除非用户明确批准其他来源或特定账户执行,否则不应使用其他端点。
安全与隐私
可能离开您机器的数据:
- - 发送到腾讯自有文档或产品网站的搜索词和页面请求
- 当官方文档不完整时,发送到用户批准的比较来源的可选备注
保留在本地数据:
- - ~/tencent/ 下的激活偏好和腾讯工作历史
- 保存在本地 Markdown 文件中的账户标签、区域默认值和决策记录
- 可信来源备注和部署约束
此技能不会:
- - 在本地 Markdown 文件中存储密钥
- 未经检查区域就声称产品在任何地方都可用
- 未经用户明确批准登录腾讯覆盖面
- 重写其自身的技能定义文件
信任
此技能依赖于腾讯自有网站以及用于验证的任何经批准的辅助来源。
仅当您信任这些服务用于您的研究查询和规划工作流程时,才安装和使用它。
范围
此技能仅:
- - 将腾讯相关请求路由到正确的产品系列
- 使用官方来源验证比较腾讯选项
- 规划腾讯云、微信生态系统和区域敏感的部署
- 在 ~/tencent/ 中保留轻量级本地记忆
此技能绝不:
- - 编造产品功能或区域可用性
- 将非官方摘要视为最终证据
- 请求密码、二维码登录令牌、Cookie 或原始密钥
- 未经用户明确批准执行更改账户的操作
相关技能
如果用户确认,使用 clawhub install
安装:
- - cloud - 通用云架构和平台权衡指导
- cloud-storage - 对象存储设计、迁移和运营模式
- payments - 支付流程设计、商户风险检查和发布规范
- in-depth-research - 更广泛的市场和来源比较工作流程
- market-research - 超越产品路由的竞争和市场层面框架
反馈
- - 如果有用:clawhub star tencent
- 保持更新:clawhub sync