RAG Anything
This MCP server provides AI assistants with comprehensive RAG (Retrieval-Augmented Generation) capabilities for processing and querying directories of documents using the raganything library with full multimodal support. Built by Jesse Merhi using Python with LightRAG for graph-based retrieval, it supports end-to-end document processing with multimodal content extraction including images, tables, equations, and text across multiple file formats (PDF, DOCX, PPTX, TXT, MD), batch processing of entire directories, and advanced querying with multiple modes (hybrid, local, global, naive, mix, bypass). The implementation features persistent RAG instances per directory for efficient re-querying, concurrent processing with configurable workers, GPT-4V integration for image analysis, and OpenAI API compatibility with custom endpoints, making it valuable for research workflows, document analysis, knowledge base construction, and building AI assistants that need advanced document understanding and retrieval capabilities without manual processing.
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
View full trust & usage report →Metadata
- Version
- 1.0.0
- Skill type
- atomic
- Execution layer
- mcp-remote
- Category
- media
- Source
- PulseMCP
- Repository
- github.com/jesse-merhi/rag-anything-mcp
- Author type
- human
- Updated
- 2026-04-25
Use via MCP
Resolve RAG Anything from your agent
Streamable HTTP transport at https://api.skillsregistry.net/mcp. No auth for read tools. Discovery: .well-known/mcp.json.
One command in your shell — Claude Code wires it up and verifies the connection. Run /mcp in any session to confirm.
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp--scope user for --scope project to commit it to .mcp.json.