$ loading_
Manage versioned golden datasets and semantic evaluations for RAG/LLM pipelines.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "golden-dataset-mcp" yet — see the docs or source repo.
Use golden-dataset-mcp to create a golden dataset with 50 QA samples and evaluate the current RAG pipeline outputs with TF-IDF cosine similarity. Return per-sample scores, the overall average score, and a list of low-scoring cases.
A structured evaluation report with sample-level scores, summary metrics, and problem cases to investigate.
Use golden-dataset-mcp to compare RAG pipeline v1 and v2 on the same golden dataset. Provide each version’s average score, the samples with the biggest differences, and identify which version performs better overall.
A comparison report highlighting overall winner and specific samples that improved or regressed.
Use golden-dataset-mcp to create a version-controlled golden dataset for our LLM application, track added, updated, and removed samples, and output the latest version summary with change notes.
A traceable dataset version history plus a current-version summary and change log.
Connect AI assistants to Discovery Engine for search, RAG, and datastore management.
Search local Markdown files and return full document contents for use.
Analyze cross-suite metrics, reports, and read-only insights across Golden Suite.
Look up Gemini API docs and get ready-to-use code snippets fast.
Access XAUUSD market data, technical analysis, backtesting, and trading risk tools.
Use Gemini Deep Research Agent for comprehensive web research and synthesis.