Index local research documents (PDF, Markdown, JSON, TXT) into Onyx RAG. Query them via onyx_ask() and onyx_search().
MIT
650,565 tools. Updated 2026-10-14 14:29
"Information on RAG Documents or Processing" matching MCP tools:
- Retrieve statistical metrics about a collection's documents, processing status, storage usage, and index health to monitor readiness and verify completion.MIT
- List documents in a Needle collection to check processing status, inventory available files, and verify document availability before searching.MIT
- Retrieve a list of all knowledge folders in your organization, each containing documents for RAG capabilities.MIT
- List RAG corpus documents, using hostname filters to locate relevant source material for editorial workflows.MIT
- Recursively ingest local directory files into the mesh's shared memory (mcl-rag) as documents, updating existing entries on re-run. Ideal for corpora, not conversational snippets.Apache 2.0
Matching MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to query OpenRouter model information including prices, ELO rankings, context, and perform comparisons.11 npm1MIT
- AlicenseNot gradedqualityDmaintenanceA production-ready Model Context Protocol server that bridges local document management with cloud synchronization (Notion) for AI agent integration, enabling seamless access and sync of local and cloud documents.12 npmMIT
Matching MCP Connectors
Authenticated Skavio document, PDF, media and transcription processing with USDC prepaid credits.
Congressional Documents — full-text search and retrieval over the official
- Retrieve document counts by processing status (pending, processing, processed, failed) to quickly answer how many documents exist in LightRAG without listing all documents.MIT
- Split raw text into fixed-size segments with configurable overlap, preparing documents for vector embeddings and RAG pipelines.MIT
- Upload 1–5 documents (PDF, text, etc.) into an app's RAG sources to make content queryable after indexing. Requires app selection or appId; async processing.MIT
- Scan local RAG source documents to detect prompt injection, secrets, and PII before indexing.MIT
- List all ingested source documents with their licenses, URLs, and processing states to review the corpus contents.MIT
- Creates a Gemini File Search store that acts as a knowledge base for RAG, returning the new store resource so you can upload documents and query them.MIT
- Analyze large documents by breaking them into chunks and processing in parallel with multiple AI models to summarize, extract information, answer questions, or search content.MIT
- Retrieve OCR processing statistics: total documents, counts by state (completed, not processed, processing, error), and per-folder breakdown.MIT
- List documents and web pages ingested into a knowledge store for RAG retrieval, with pagination and filtering.MIT
- Search uploaded documents using RAG to find answers with citations. Ask questions to retrieve information from your knowledge base.MIT
- Find relevant documents in the RAG system using semantic search with customizable similarity thresholds and result limits.MIT
- Scans document text, source, and type to validate and prepare content for retrieval-augmented generation (RAG) ingestion.Apache 2.0
- Analyze document content using AI to bulk assign relevant tags. Skip already tagged documents and control processing volume.-
- Retrieve a list of all documents currently indexed by the RAG service.-
- Create a named local vector index for retrieval-augmented generation. Documents added are embedded via Ollama for local RAG without cloud dependencies.-