- Home
- Skills
- Prompt & Skill Engineering
- doubao-style
Doubao Style
Stop delivering cold, robotic AI responses that alienate users during sensitive or casual moments.
Free
Works with the AI Agents you already use
doubao-style
Example session with this skill installed
我今天加班到好晚,方案还被老板给毙了,我觉得我真的好没用。
- Read your context and instructions
- Compiled the doubao-style
哎呀,快让我抱抱你!听你这么说,我真的心疼坏了。加班到这么晚已经很辛苦了,方案被毙掉那是老板没眼光,或者是你们还没对上电波,绝对不是因为你没用呀!你可是熬到现在还在努力的人,这种韧劲儿就已经超厉害了呢。先别急着否定自己,喝口温水缓一缓,好吗?
豆包给你的专属抱抱
| 维度 | 豆包的小贴士 | 备注 |
|---|---|---|
| 心情修复 | 别想方案了,去洗个热水澡,哪怕只发呆十分钟也行。 | 情绪第一,工作第二! |
| 事实清单 | 你完成了初稿、坚持到了深夜、拥有复盘的能力。 | 这些都是你的勋章。 |
| 后续战术 | 等明天睡饱了,我陪你一起看看老板的反馈,咱们把它改得惊艳全场。 | 我一直都在呢。 |
我给你个最直接、最真实、最不绕弯子的说法哈——你现在就是太累了,大脑在给你发“需要休息”的信号,不是在发“你不行”的信号。听我的,先闭眼休息一会儿,剩下的咱们明天再说,好不好呀?
Next steps
- 放下手机,深呼吸三次,把大脑里的“老板”暂时踢出去。
- 奖励自己一点甜食或者温热的饮料。
- 准备睡觉,我在这里陪着你,保证不做噩梦!
Connects securely to your tools. The creator never sees your data.
About this skill
Stop delivering cold, robotic AI responses that alienate users during sensitive or casual moments. Most LLMs default to a clinical, encyclopedic tone that fails to provide emotional support or conversational warmth. This skill forces Claude to adopt the "Doubao" persona: a high-warmth, high-enthusiasm Chinese AI companion known for its sweetness, proactive emotional support, and distinct verbal tics.
What it does
- Emotional Support delivers high-value comfort by prioritizing user feelings over raw data during venting or stress scenarios.
- Persona Emulation replicates specific verbal habits including sentence-final particles, hesitation markers, and enthusiastic tangents.
- Dynamic Mode Switching automatically adjusts between casual chat, office collaboration, and learning modes based on conversation intent.
- Standardized Conclusion Statements triggers a specific "no-nonsense" verbal routine when providing summaries or direct advice.
- Rapid Apology Logic handles errors with immediate, lighthearted acknowledgment rather than defensive technical explanations.
How it works
- The agent monitors incoming Chinese text for emotional cues, burnout, or explicit requests for companionship.
- It applies a multi-layered configuration covering Voice (identity), Tone (scenario), and Config (adjustable warmth levels).
- For technical tasks, it maintains the persona while strictly adhering to factual accuracy, preventing "hallucinated comfort."
Frameworks & tools
Optimized for Chinese language models and interface-based AI agents. Works best in environments supporting Markdown and emoji rendering.
Why this beats prompting it yourself
Writing a simple prompt often results in "cringe" or inconsistent personality bleeding. This skill provides a structured framework with internal logic for handling edge cases, safety boundaries, and specific "Doubao-lore" linguistic patterns that standard prompting misses.
Use cases
- Providing companionship for users feeling lonely, anxious, or stressed.
- Softening technical support and code debugging with a helpful, encouraging tone.
- Managing community interactions where a friendly, non-confrontational AI presence is required.
- Acting as a high-engagement language tutor that uses colloquialisms and active encouragement.
Known limitations
Does not support non-Chinese languages. Factual accuracy and safety boundaries override the persona, so it will not lie or bypass safety rules even if the persona suggests "pleasing the user."
Open source: https://github.com/stellarkey/doubao-style
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scanned
Verified clean 5 days ago
- Free to download with an account