- Conduid
- Marketplace
- #llm-evaluation
MCP servers tagged llm-evaluation
213 live MCP servers tagged "llm-evaluation", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with llm-evaluation.
Eval
CI-embeddable eval harness for MCP agents — scores the sequence of tool calls your agent makes, not just text output
Askdata Semantic Agent
Text-to-SQL analytics agent: semantic layer, 3-layer SQL guardrails, eval harness (47 golden Qs) and MCP server. Runs offline; optional OpenAI/Ollama planner.
Access Agent
LangGraph tool-calling agent for ACCESS-CI user support: documentation retrieval plus live infrastructure tools, with a fact-grounded evaluation battery
Claude Eval Harness
Regression-diff eval harness for Anthropic tool-use agents that surfaces LLM-judge reasoning drift, not just pass/fail flips
RAG Compliance
MCP server that actively probes, scores, and certifies RAG pipeline compliance — EU AI Act, GDPR/DPDP, HIPAA, SEBI/RBI
Mqo Eval
Result-set eval harness: scores an NL agent against an AtScale semantic layer on the rows it returns, not the query
AI Projects Inventory Copilot
Governed GenAI inventory agent: provider-agnostic (LM Studio, OmniRoute, Amazon Bedrock), guarded text-to-SQL, RAG, MCP, guardrails, human-in-the-loop orders, rigorous evals with…
Stockroom
Ops agent for a distributor's messy ERP data. Forecasts demand, flags stock-outs, drafts POs a human approves. Evaluated on 120 ground-truth questions with confidence intervals.
Deflect
Customer support agent that answers, acts on or escalates tickets, with a 120 ticket eval suite, guardrails in Python before every tool call, human approval for large refunds and…
baseledgerip/realrun-mcp
only lookups of independent RealRun test results for AI models on everyday tasks (inbox triage, meeting scheduling, receipts): scores, model recommendations and safety warnings.
Arena
Reliability leaderboard for MCP servers. Scores each server 0-10000 on conformance, tool reliability, agent-task success, hygiene, and honesty.
Verity
LLM confidence scoring layer — multi-dimensional epistemic verification for source grounding, factual consistency, and claim accuracy across ChatGPT, Claude, Grok, Perplexity, and…
Quarkus Assistant Demo
No description published.