Marketplace for trusted AI agent skills

Skill Detail

Connect agents to OpenRAG knowledge bases over MCP

Use OpenRAG's built-in MCP endpoint so agents can ingest documents, search a knowledge base, create filters, and run RAG-backed chat against a deployed OpenRAG instance.

Integrations & ConnectorsMCP
Integrations & Connectors MCP Security Reviewed
⭐ 4.4k GitHub stars ⬇ 52/wk npm
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill connect-agents-to-openrag-knowledge-bases-over-mcp Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
View source Documentation
At a glance
Tools required
OpenRAG deployment; OpenRAG API key; MCP client with streamable HTTP support such as Cursor, Claude Desktop, IBM Bob, or another MCP-capable agent runtime
Install & setup
Install and run OpenRAG from https://docs.openr.ag/quickstart or https://docs.openr.ag/docker, create an OpenRAG API key, then add an MCP server pointing to http://localhost:3000/mcp or the deployed /mcp URL with the X-API-Key header.
Author
Langflow AI
Publisher
Organization
Last updated
Jul 22, 2026
Quick brief

Use this skill when an agent needs to operate an OpenRAG knowledge base instead of asking a human to use the OpenRAG UI manually. The operator runs OpenRAG, creates an API key, connects an MCP-capable client to the `/mcp` endpoint, and lets the agent ingest files or URLs, search semantically, chat against scoped knowledge, check ingestion tasks, and manage knowledge filters.

How it works

What this skill actually does

This is not a generic RAG platform listing. The scope is the repeatable MCP workflow around OpenRAG’s documented tools: `openrag_ingest`, `openrag_search`, `openrag_chat`, task-status checks, settings/model inspection, and knowledge-filter management. Use it when the agent needs a local or hosted retrieval layer it can call directly during research, support, documentation, or enterprise search workflows.

Install OpenRAG using the upstream quickstart or Docker documentation, then configure an MCP client with a streamable HTTP server URL such as `http://localhost:3000/mcp` and an `X-API-Key` header. The deprecated `openrag-mcp` package should not be used; upstream documents the built-in HTTP endpoint as the supported path.