wiki-embed-qdrant

To embed wiki chunks into a Qdrant collection, this skill converts semantic chunks into 768-dim Gemini embeddings and upserts them for fast retrieval. You run the embed_chunks.py script after configuring your Gemini and Qdrant credentials.

Indexing wiki content for semantic search in Qdrant Cloud involves generating Gemini vector embeddings from text chunks and upserting them into a collection. This skill automates that pipeline, including batch processing to handle large volumes efficiently.

Yes, you need both Gemini and Qdrant credentials to generate vector embeddings for wiki text. Gemini provides the 768-dimensional embedding generation, while Qdrant Cloud hosts the collection where these embeddings are upserted for retrieval.

Yes, you can resume vector embedding jobs in Qdrant if interrupted. This skill automatically resumes from the last processed chunk, ensuring long-running batch embedding jobs recover without duplicating previously upserted data.

Semantic chunking with Gemini embeddings splits wiki content into meaningful segments, converts them to 768-dimensional vectors, and upserts them into Qdrant. This allows a production RAG backend to perform fast semantic search over the wiki knowledge base.

Upserting Gemini embeddings into Qdrant Cloud requires specific Python dependencies to handle API connections and batch processing. You must install these dependencies alongside configuring your Gemini and Qdrant credentials before running the embedding script.