← Back to AI Insights
Gemini Executive Synthesis

This issue is a feature request aggregation for the "ex-skill" project. Users are requesting advanced AI capabilities such as voice cloning, multi-ex support, scenario simulation, session summarization for continuous memory, and more realistic interaction patterns (e.g., delayed responses, "breaking up" probability). There are also mentions of data extraction and file size/type issues.

Technical Positioning
The project aims to create a highly realistic and emotionally resonant AI simulation of an ex-partner. The requested features push towards greater immersion, personalization, and advanced AI conversational capabilities, moving beyond basic text-based imitation.
SaaS Insight & Market Implications
This issue reveals a strong user demand for sophisticated AI features within the `ex-skill` product. Users are pushing for enhanced realism and utility, including voice cloning, multi-ex support, scenario simulation, and continuous memory through session summarization. Critically, requests for non-instantaneous, human-like response delays and even a "breakup" probability indicate a desire for highly nuanced, emotionally authentic AI interactions. These features move the product beyond simple text generation towards a more immersive and psychologically complex simulation. Addressing these demands requires significant advancements in AI capabilities, particularly in multimodal AI and complex conversational state management, to meet user expectations for a truly convincing digital companion.
Proprietary Technical Taxonomy
AI Skill 关系反思与经验教训功能 主动分享功能 小红书帖子 聊天记录提取问题 文件大小和类别问题 语音克隆功能 切换多个前任

Raw Developer Origin & Technical Request

Source Icon GitHub Issue Apr 2, 2026
Repo: therealXiaomanChu/ex-skill
欢迎大家在本条下面评论任何需要改进或者想要加入的功能 ˆ ̳◝ ·̫ ◜ ̳ˆ

@W1ndys (github.com/W1ndys
加入了#增加关系反思与经验教训功能
@fizherbeart (github.com/therealXiaomanChu...
# 添加主动分享功能,模仿前任分享一些小红书帖子
@UniUni2000:聊天记录提取问题
@buchenzhang: 文件大小和类别问题

Developer Debate & Comments

enpyth • Apr 2, 2026
1. 语音克隆功能,提供ta的音频后能克隆语音,模仿发声 2. 支持切换多个前任 3. 设置场景,看多个前任的对话过程
masachi • Apr 2, 2026
希望能加上 每一次对话之后/对话特定条数 能输出 session summary 然后在开始新对话前 加载前N天/前N个summary 实现记忆不中断? 我请gpt写了一下可能可以实现的方案: https://chatgpt.com/share/69ce03e5-df74-83e8-a46e-8b61a27b3edc
zerokileom • Apr 3, 2026
1%的概率下分手,成为前任
ZUOXIANGE • Apr 3, 2026
不要秒回
nannan119 • Apr 3, 2026
@ZUOXIANGE 你小子,轮舔还是你:laughing: 训练时,提示词直接给他人设要求他30s-12s内随机时间回复就行了

Adjacent Repository Pain Points

Other highly discussed features and pain points extracted from therealXiaomanChu/ex-skill.

Extracted Positioning
The user asks if it's possible to "replicate" an ex-partner via the AI skill even if all records and contact information have been deleted. This is another instance of the data availability problem, compounded by the emotional context of complete data loss. The comments are humorous but reinforce the idea of "replication."
The product's core value is "distilling" an ex-partner. This issue highlights the extreme end of the data scarcity problem, where the user has no input data.
Top Replies
chensanle • Apr 1, 2026
她不是已经复刻一个了吗
liubo-it • Apr 1, 2026
> 她不是已经复刻一个了吗 6666
CccryQwQ • Apr 1, 2026
> 她不是已经复刻一个了吗 666
Extracted Positioning
The user raises a profound psychological and ethical concern: if the AI is too realistic, it might replicate the negative aspects of the past relationship, specifically the act of "abandonment." This points to the emotional risks and ethical considerations of highly realistic AI simulations.
The product aims for high fidelity in simulating an ex-partner. This issue highlights the double-edged sword of such realism, where negative experiences could also be replicated.
Top Replies
1AKO1 • Apr 1, 2026
呜呜
Bit-urd • Apr 1, 2026
我有个朋友让你憋说了😠
dangyuhang0825 • Apr 1, 2026
呜呜呜呜,我要举报你!呜呜呜
Extracted Positioning
This issue is a reflective piece on human memory and its selective nature, contrasting it with the idea of an AI skill. The user shares a 40,000-word memoir, implying that such detailed, emotionally rich narratives could serve as input for the "ex-skill" AI. This highlights the potential for qualitative, narrative data as a training source.
The product aims to capture the essence of an ex-partner. This issue suggests that deeply personal, narrative accounts, beyond just chat logs, could be valuable for achieving a more nuanced and emotionally accurate AI representation.
Top Replies
poboll • Apr 1, 2026
看到你这句留言,心里忽然有些发酸。 其实大脑是很吝啬的,它每天都在清理平庸的日常,却唯独对某些毫无逻辑的瞬间网开一面。 您也是很感性的人儿呀,一定是很热爱生活的人吧。 岁月虽然一直推着人往前走,但只要...
wdkang123 • Apr 1, 2026
> 看到你这句留言,心里忽然有些发酸。 > > 其实大脑是很吝啬的,它每天都在清理平庸的日常,却唯独对某些毫无逻辑的瞬间网开一面。 > > 您也是很感性的人儿呀,一定是很热爱生活的人吧。 > > 岁月虽然一直推着人...
wdkang123 • Apr 1, 2026
![Image](https://github.com/user-attachments/assets/c36fe4dd-2491-45fc-a0e5-a893437e0128) ![Image](https://github.com/user-attachments/assets/b764b56e-f05b-4138-b014-c94d43fb400b)
Extracted Positioning
The user expresses a desire to "distill the physical body" and replace the "head" (intelligence/personality) with advanced LLMs like Opus or Grok, implying dissatisfaction with the current AI's cognitive capabilities or a desire for a different kind of simulation. This is a feature request for modularity and advanced AI integration.
The product aims to "distill an ex-partner into an AI Skill." This user's comment suggests a desire to separate the "essence" (personality/communication style) from the underlying intelligence, or to upgrade the intelligence with state-of-the-art models.
Extracted Positioning
The user is asking about the data handling capacity of the "ex-skill" product, specifically if it can process "tens of gigabytes" of data. This points to a technical limitation or concern regarding data ingestion and storage for training the AI.
The product relies on user-provided data to "distill" an ex-partner. The ability to handle large datasets is crucial for the fidelity and depth of the AI model.

Engagement Signals

5
Replies
open
Issue Status

Cross-Market Term Frequency

Quantifies the cross-market adoption of foundational terms like AI Skill and session summary by tracking occurrence frequency across active SaaS architectures and enterprise developer debates.