# MiniRAG

> Use this tool when you need to efficiently retrieve and generate text based on specific entities and relationships, or when you want to augment language model capabilities with reliable and structured output. MiniRAG solves problems related to knowledge retrieval and generation by providing two approaches: naive vector database querying and smart graph-based retrieval. It takes in queries and returns answers with cited sources in JSON format, making it ideal for use cases involving small language models and optimized knowledge retrieval workflows.

Canonical page: https://skillsregistry.net/skills/witwicki-minirag  
JSON: https://api.skillsregistry.net/v1/skills/witwicki-minirag

## Description

An MCP server wrapper around MiniRAG that provides efficient retrieval-augmented generation capabilities through two distinct approaches: naive vector database querying and smart graph-based retrieval using entity relationships and reasoning paths. Leverages client-managed LLM inference with structured output constraints for enhanced reliability, while simplifying MiniRAG setup through UV dependency management. Supports configurable embedding providers including Ollama and OpenAI-compatible services, returning answers with cited sources in JSON format for knowledge retrieval workflows optimized for small language models.

## Trust

- **Trust score (0–1):** 0.65
- **Verification tier:** scanned
- **Last scanned:** 2026-09-01

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** database
- **Updated:** 2026-09-01

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/witwicki-minirag)
- **Repository:** <https://github.com/witwicki/minirag-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "witwicki-minirag"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/witwicki-minirag` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/witwicki-minirag/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt