ragscore

IntegrationDatabases & data

ragscore helps you find out how well your AI answers questions when it looks things up in your documents. It creates question-and-answer test datasets, scores your retrieval-augmented generation system's answers, and pinpoints why answers fail. It works with any large language model.

This server has no hosted endpoint yet, so ahel can't serve it. You can still add it. It stays paused until ahel can serve it.

After adding ragscore, see its repository at github.com/hzyai/ragscore for setup details. Start by generating a test dataset, then run an evaluation to see where your system's answers fall short.

What your AI can do with it

  • Generate question-and-answer datasets for testing
  • Evaluate how accurately your RAG system answers
  • Diagnose the causes of failed answers
  • Run evaluations with any large language model

Signals

GitHub stars
15
Forks
2
Last commit
May 2026
Advanced
Delivery
ragscore MCP server → your ahel gateway (mcp.ahel.ai) → every connected AI client.
Catalog kind
mcp-server
Gateway key
io-github-hzyai-ragscore
Source
github.com/hzyai/ragscore