What problem does it solve? Setting up retrieval-augmented generation usually requires building embedding pipelines and managing vector infrastructure. This Skill lets you create Pinecone Assistants, upload documents, and ask questions with cited answers directly from Claude Code, with no RAG plumbing required. ## Core Features & Use Cases - Assistant Lifecycle Management: Create assistants with custom instructions and region selection, and list existing assistants with file details. - Document Upload & Incremental Sync: Upload Markdown, PDF, DOCX, TXT, and JSON files, then keep assistants in sync with local folders using mtime/size change detection, dry-run previews, and optional deletion of removed files. - Cited Chat & Context Retrieval: Ask questions and receive answers with source citations, or pull raw context snippets with relevance scores for custom RAG workflows. - Use Case: Point the assistant at your ./docs folder, sync it after every git pull, and ask natural-language questions like "what are the main features?" to get answers grounded in your own documentation. ## Quick Start Create a Pinecone assistant called docs-bot, upload my ./docs folder, and ask it what the main features are.
