Architect high-precision retrieval augmented generation pipelines. These specialized subagents guide document parsing, semantic chunking, embedding generation, and vector index provisioning.
4 Verified
Claude, Cursor, Windsurf
1-Click Install
Weekly Benchmarks
Ranked by real-world developer adoption, benchmarked reliability, security posture, and documentation depth.
Document chunking & vector preparation
npx -y skills add jgamaraalv/delivery-loop --skill doc-to-vector-dataset-generator --agent claude-codeEmbedding workflows & vector indexing
npx -y skills add jgamaraalv/delivery-loop --skill embedding-pipeline-builder --agent claude-codeEnd-to-end RAG architecture
npx -y skills add jgamaraalv/delivery-loop --skill rag-pipeline-builder --agent claude-codeVector database deployment & tuning
npx -y skills add jgamaraalv/delivery-loop --skill vector-db-setup --agent claude-code| Rank | Name | Stars / Popularity | Best Use Case | Details |
|---|---|---|---|---|
| #1 | Doc-to-Vector Generator | Curated Pick | Ask your agent: "Use /doc-to-vector-dataset-generator to execute and coordinate this task." | View → |
| #2 | Embedding Pipeline Builder | Curated Pick | Ask your agent: "Use /embedding-pipeline-builder to execute and coordinate this task." | View → |
| #3 | RAG Pipeline Builder | Curated Pick | Ask your agent: "Use /rag-pipeline-builder to execute and coordinate this task." | View → |
| #4 | Vector Database Setup | Curated Pick | Ask your agent: "Use /vector-db-setup to execute and coordinate this task." | View → |
Our evaluation methodology ensures every listed tool satisfies production engineering standards.
High precision-at-k and recall through semantic re-ranking and metadata filtering.
Decoupled stages for chunking, embedding, vector storage, and reranking.
The RAG Pipeline Builder focuses on query-time retrieval, re-ranking, and prompt augmentation, while the Embedding Pipeline Builder focuses on batch ingestion, token chunking, and vector index creation.