A
licenseNot graded
qualityB
maintenanceEnables agents to run evals mid-task, providing tools to validate suites, execute model comparisons, and inspect calibration and win-rate reports.
MIT
Coval MCP Server
Officialby coval-ai
Related Servers
Alternatives to Coval MCP Server
No user-submitted related servers found.
Related Servers
- AlicenseAqualityAmaintenanceProvides read-only tools to generate test plans for AI agents, score agent runs, and produce reusable evaluation scorecards.3MIT
- AlicenseNot gradedqualityBmaintenanceEnables LLM evaluation and observability by uploading documents, building test sets, running RAG pipelines, and automatically scoring answers for groundedness, hallucination risk, retrieval quality, latency, and cost, with tools exposed to MCP-compatible clients.1MIT
VerifyAX MCPofficial
AlicenseAqualityAmaintenanceEnables conversational access to the VerifyAX agent-evaluation platform, exposing tools for agent evaluation workflows through natural language.121Apache 2.0- AlicenseNot gradedqualityCmaintenanceEnables AI agents to access observability and evaluation data, including run history, span traces, LLM-as-judge evaluation results, and regression reports.MIT
- AlicenseNot gradedqualityDmaintenanceEnables automated LLM red teaming by submitting asynchronous test runs, retrieving aggregated metrics, and accessing artifacts.4CC BY-SA 4.0
TDQS
A3.7/5.0
Scored across 18 tools
Disambiguation5/5
Each tool targets a distinct resource (agent, run, test case, test set, metric, persona) with specific CRUD operations, leaving no ambiguity between tools.
Naming Consistency5/5
All tool names follow a consistent verb_noun pattern (create_, get_, list_, update_), making it easy to predict tool purpose from the name.
Tool Count5/5
18 tools cover the complete lifecycle of evaluating AI agents (agents, personas, test sets, runs, metrics) without being overly numerous or sparse.
Completeness4/5
Core workflows are covered, but missing delete operations for agents, test cases, and test sets, as well as update for test sets, leave minor gaps in the surface.
Maintenance
ActivityActive
ResponsivenessNo issues