skills.rest
hhhh124hhhh
hhhh124hhhhCOMMUNITY
Explore all skills by this author →
GitHub

ai-partner-chat

Retrieve top-k vectorized notes with BAAI/bge-m3 embeddings from ChromaDB for persona-driven chat responses.

5|1|Updated Nov 11, 2025
Search Tags:#knowledge base#vector search#contextual AI#persona management#personalized chat#note retrieval#AI partner
One-click install
npx skills add https://github.com/hhhh124hhhh/LangGraph-Partner --skill ai-partner-chat
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-partner-chat
Source: https://github.com/hhhh124hhhh/LangGraph-Partner/tree/main/.claude/ai-partner-chat
Command: npx skills add https://github.com/hhhh124hhhh/LangGraph-Partner --skill ai-partner-chat

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires chromadb, sentence-transformers, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

This Skill eliminates generic AI interactions by providing a personalized AI partner that remembers your context, preferences, and past notes. It ensures every conversation is relevant, in-depth, and tailored to your unique knowledge base, saving you time from repeating yourself or providing background information.

Core Features & Use Cases

  • Personalized Conversations: Integrates user and AI personas to adapt communication style and tone, creating a more natural and effective dialogue.
  • Context-Aware Responses: Utilizes vectorized personal notes to retrieve and reference your previous thoughts, ideas, and knowledge, ensuring coherent and informed replies.
  • AI Agent Chunking: Dynamically analyzes the format of your notes (e.g., daily logs, structured documents) and generates optimal chunking strategies on-the-fly for precise information retrieval.
  • Use Case: Imagine you're a researcher with extensive notes. Your AI partner can recall specific insights from your past work, suggest connections between ideas, and help you brainstorm, all while maintaining a consistent understanding of your ongoing projects and learning style.

Quick Start

  1. Set up your Python environment and install dependencies.
  2. Create config/user-persona.md and config/ai-persona.md from the provided templates.
  3. Place your Markdown notes in the notes/ directory.
  4. Initialize the vector database by running python scripts/chunk_and_index.py.
  5. Start a personalized conversation with your AI partner, asking questions or discussing topics, and observe how it references your notes and persona.

Frequently Asked Questions about ai-partner-chat

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a personalized AI chat that remembers my notes and context?▼

Personalized AI chat integrates your notes into a vector database using embeddings, so the AI retrieves relevant past ideas when responding. This Skill uses ChromaDB and sentence-transformers to chunk your Markdown notes, embed them with BAAI/bge-m3, and inject retrieved context into responses shaped by your user and AI personas.

What's the difference between vector search and keyword search for retrieving personal notes?▼

Vector search finds semantically similar notes by converting text to embeddings, capturing meaning beyond exact word matches. Keyword search only matches phrases. Vector search excels at surfacing related ideas from your knowledge base even when phrased differently, enabling context-aware AI responses.

Can I use ChromaDB and sentence-transformers to create a knowledge base for AI conversations?▼

Yes. ChromaDB stores and indexes embeddings generated by sentence-transformers, forming a vector database. This Skill demonstrates the full pipeline: chunk your notes dynamically, generate embeddings, store them in ChromaDB, and retrieve top-k results to ground AI responses in your personal knowledge.

How do I set up user and AI personas to shape conversation tone?▼

Create Markdown files defining your communication style, goals, and preferences (user persona) and the AI's tone, expertise, and response approach (AI persona). The Skill uses these templates to generate responses consistent with both personas, ensuring natural, tailored dialogue aligned with your needs.

What happens if my notes are in different formats—can the AI handle mixed note structures?▼

This Skill includes AI-driven chunking that analyzes your note format on-the-fly, detecting patterns in daily logs, structured documents, or mixed styles. It generates optimal chunking strategies per format, ensuring precise retrieval and indexing regardless of how you organize your notes.

Do I need machine learning expertise to deploy vector search for my personal notes?▼

No. This Skill abstracts embedding generation and vector indexing; you provide Markdown notes and personas, run the initialization script, and start chatting. Sentence-transformers and ChromaDB handle the ML infrastructure, requiring only Python and the listed dependencies.

Related Skills

Explore More Personal & Entrepreneur Skills→
wyhlovecppwyhlovecpp

memory-system

Index Markdown files into vector and full-text search with SQLite.

Community
Intermediate
MykalMachonMykalMachon
3

researcher

Research topics and generate structured summaries with cited references.

Community
Intermediate
RheasilviaRheasilvia

chroma

Manage vector embeddings and perform semantic similarity searches in chromadb databases.

Community
Intermediate
aisa-groupaisa-group
90

rag-implementation

Build retrieval-augmented generation systems with embeddings, vector stores, and reranking.

Official
Intermediate