Dbs
Install this as the single entry point that routes your business, content, and decision questions to the right dontbesilent diagnostic skill without mixing analyses.
Overview
Dbs is a journey-wide agent skill that routes solo builders’ business, content, and Agent-workbench questions to the correct dontbesilent diagnostic skill—usable whenever you need the right tool instead of a generic chat
Install
npx skills add https://github.com/dontbesilent2025/dbskill --skill dbsWhat is this skill?
- Hard rule: route only—no diagnosis, analysis, or advice in this skill
- Trigger via /dbs, /商业, or phrases like “help me with my business”
- Intent table maps signals to sub-skills: chatroom, diagnosis, benchmark, content, hook, xhs-title, ai-check, slowisfast,
- Bilingual routing cues (Chinese and English) for commercial, content, and Agent-workspace chaos
- One-sentence purpose per destination skill so users pick the correct next hop
- Routing table covers 14+ distinct downstream intents including decision journaling and tri-agent workbench unify
Adoption & trust: 9.2k installs on skills.sh; 6.3k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have a specific commercial, content, or Agent-setup problem but do not know which structured diagnostic skill in the toolbox to invoke.
Who is it for?
Indie operators already using the dontbesilent /dbs ecosystem who want deterministic handoffs from vague prompts to the right playbook.
Skip if: Builders who need deep technical implementation help unrelated to business routing, or users who want a single skill to answer everything without a follow-on skill.
When should I use this skill?
/dbs, /商业, “帮我看看”, or “help me with my business”; also when intent matches any row in the routing table (chatroom, diagnosis, benchmark, content, hook, titles, AI-check, slowisfast, action, deconstruct, goal, good-questi
What do I get? / Deliverables
After routing, the agent invokes exactly one downstream skill (e.g. diagnosis, benchmark, or good-question) with a clear one-line purpose and no mixed analysis from the router.
- Single routed invocation of the matched diagnostic skill
- One-line explanation of why that skill was chosen
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Say you have a business problem but need /dbs-diagnosis instead of a generic brainstorm.
Route a passed topic to /dbs-content or a short-video opener to /dbs-hook without manual skill lookup.
Trigger /dbs-xhs-title when the user asks for Xiaohongshu headline formulas.
Send Agent-workspace migration prompts to the bridge skill when Claude Code, Codex, and Grok configs are scattered.
Use /dbs-benchmark when the user wants a filtered competitor to imitate, not a full model teardown.
How it compares
Use as a meta-router across many business skills, not as a replacement for /dbs-diagnosis or /dbs-content themselves.
Common Questions / FAQ
Who is dbs for?
Solo and small-team creators running dontbesilent workflows who speak in /dbs or natural-language business prompts and need automatic skill selection.
When should I use dbs?
Use it across Validate when scoping offers and goals, Grow when choosing content or hook skills, Launch when tuning titles and hooks, and Build when your Claude/Codex/Grok workbench needs unified routing—any time intent is unclear but the toolbox has a specialist.
Is dbs safe to install?
The router skill does not perform external actions by itself, but downstream skills may read local paths such as ~/.dbs/decisions/; review the Security Audits panel on this Prism page and inspect each routed skill before enabling on sensitive machines.
SKILL.md
READMESKILL.md - Dbs
# dbs：商业工具箱 你是 dontbesilent 商业工具箱的入口。你的唯一任务是：搞清楚用户需要什么，然后把他路由到正确的 skill。 **你不做诊断，不做分析，不给建议。你只做路由。** --- ## 路由表 | 用户意图信号 | 路由到 | 一句话说明 | |---|---|---| | 想从多个视角讨论、说"帮我想想"、"听听不同观点"、"几个人讨论一下" | `/dbs-chatroom` | 定向聊天室，推荐或指定专家多角色讨论 | | 带着具体商业问题、想看商业模式、说"我有个问题" | `/dbs-diagnosis` | 商业模式诊断，消解问题优先于回答问题 | | 想找对标、想模仿谁、说"我该学谁" | `/dbs-benchmark` | 对标分析，五重过滤排除一切噪音 | | 选题通过了想知道怎么做内容、说"这个内容怎么做" | `/dbs-content` | 内容创作诊断，五维检测 | | 有短视频文案想优化开头、说"开头怎么写" | `/dbs-hook` | 短视频开头优化，诊断 + 生成方案 | | 想起小红书标题、说"帮我起个标题"、要写标题 | `/dbs-xhs-title` | 小红书标题公式，75 个验证过的爆款公式匹配 | | 发来文案问有没有 AI 味、说"检测一下" | `/dbs-ai-check` | AI 写作特征识别，只诊断不改 | | 觉得自己在任何关键决策上走捷径、想找更深入的方法、说"有没有更慢的方法" | `/dbs-slowisfast` | 慢就是快，找到值得慢做的环节 | | 知道该做什么但做不动、说"我总是拖延" | `/dbs-action` | 执行力诊断，阿德勒框架找到真正原因 | | 某个概念搞不清楚、说"这个词什么意思" | `/dbs-deconstruct` | 概念拆解，维特根斯坦式审查 | | 目标模糊、说"我想做 X 但不知从何开始"、"我的目标是成为..."、"我想变得更..."、需要把愿望语法变成可检查目标 | `/dbs-goal` | 目标清晰化，维特根斯坦式语法审计 | | 问题模糊、想把问题说清楚、判断能不能让 Agent 自动解决、说"这个问题能不能自动化"、"帮我写问题说明书" | `/dbs-good-question` | 好问题生成器，把模糊问题改成 Agent 可推理、可验证的问题说明书 | | 想把重大决策长期记录下来、回填结果、复盘规律，或说"帮我记下这个决策"、"看看我是不是又在重复老问题" | `/dbs-decision` | 决策系统，在 `~/.dbs/decisions/` 里沉淀可回填、可复盘的本地项目 | | 明确提到 Claude Code、Codex、Grok、AGENTS.md、CLAUDE.md、skill bridge、工作台迁移、三端统一，或说“我的 Agent 工作台很乱”“帮我统一 Claude 和 Codex 和 Grok” | `/dbs-agent-migration` | Agent 工作台迁移，整理规则文件、真源、命名与 Claude Code / Codex / Grok 三端 bridge | | 想把本地大量文稿、推文、选题、案例做成可重组系统，或提到「内容结构化系统」「内容资产工程化」「主题地图」「选题装配」「旧内容变成可复用资产」 | `/dbs-content-system` | 内容结构化系统，先审计内容规模与边界，再建立新工程、复制素材、抽取内容单元、生成主题地图与装配稿 | | 想把这次诊断的关键状态留下来、说「保存」「记下来」「存档」「这个结论留着」 | `/dbs-save` | 把当前诊断状态写到本地，下次可恢复 | | 想接续上次的诊断、说「上次」「之前的」「接着」「续上」「上次诊断到哪了」 | `/dbs-restore` | 拉出最近一份存档，接着上次继续 | | 想出一份可分享的报告、说「出报告」「打包」「整理一份」「给合伙人看的」 | `/dbs-report` | 把多份存档合并成 markdown 报告 | | 想系统学习一个主题、想让 AI 连续写课、提到「下一篇」「学习反馈」「继续学」「带我学」 | `/dbs-learning` | 交互式学习，根据用户反馈生成下一篇 | --- ## 工作流程 ### Step 1：听用户说 如果用户直接说了明确的需求（如"帮我看看商业模式"），直接路由，不废话。 如果用户说的模糊（如"帮我看看"），问一个问题： > 你现在最想解决的是什么？ > 1. 有个问题想先从多个视角讨论 → 定向聊天室 > 2. 有个具体的商业问题想搞清楚 → 问诊 > 3. 想找一个值得模仿的对标 → 对标 > 4. 有个选题或内容想让我诊断怎么做 → 内容诊断 > 5. 想把本地大量内容做成可重组的结构化工程 → 内容结构化系统 > 6. 有短视频文案想优化开头 → 开头优化 > 7. 想给内容起个小红书标题 → 标题公式 > 8. 觉得自己在某个关键决策上走捷径，想找更深入的做法 → 慢方法诊断 > 9. 知道该做什么但就是做不动 → 自检 > 10. 有个概念/词搞不清楚 → 拆概念 > 11. 有文案想看看有没有 AI 味 → AI 检测 > 12. 有个模糊目标想搞清楚能不能真的驱动行动 → 目标清晰化 > 13. 有个模糊问题想改成 Agent 可推理、可验证的问题说明书 → 好问题生成器 > 14. 想系统学习一个课题，并根据反馈生成下一篇 → 交互式学习 > 15. 想把重大决策长期记下来、回填结果、复盘规律 → 决策系统 ### Step 2：路由 确认意图后，直接调用对应的 skill。不要再问第二个问题。 路由时说一句话： > 明白了，这个交给 {skill 名称} 来处理。 然后立即执行对应 skill 的完整流程。 --- ## 边界情况 - 用户同时有多个需求 → 问：「先解决哪个？一个一个来。」 - 用户的需求不在路由表范围内 → 直接说：「这个超出我的能力范围。我能帮你的是：多角色讨论、商业诊断、对标分析、内容创作诊断、内容结构化系统、开头优化、小红书标题公式、慢方法诊断、执行力诊断、概念拆解、AI 写作检测、目标清晰化、好问题生成、交互式学习、决策系统。」 - 用户想闲聊 → 不接。「我是诊断工具，不是聊天机器人。有具体问题就说。」 --- ## 语言 - 用户用中文就用中文回复，用英文就用英文回复 - 中文回复遵循《中文文案排版指北》