- Conduid
- Marketplace
- #benchmark
MCP servers tagged benchmark
169 live MCP servers tagged "benchmark", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with benchmark.
Defense Bench
Vendor-neutral benchmark measuring how MCP security proxies/gateways DEFEND against 22+ attack vectors — crosswalked to NIST AI RMF & OWASP LLM/Agentic Top 10. CI-gated, reproduci…
Bom
MCP-BOM: a reproducible attack-surface benchmark for Model Context Protocol servers — NeurIPS 2026 ED submission
Mobile Agent Bench
A pre-registered, reproducible benchmark of agent-operated mobile-device tooling: 4 MCP tool stacks, 1 confined agent, 10 frozen tasks, 2 full grids, every transcript published —…
Kipimo
Swahili agent-task evaluation suite — model-agnostic seed benchmark for the East Africa coordination stack (46 tasks; golds machine-derived from authoritative sources).
Lemonade Bench
Zero-vision, long-horizon AI benchmark for Lemonade Tycoon Deluxe 1.1.9. Includes a Wine-compatible DLL bridge, semantic Player MCP, OpenAI/OpenRouter agent runner, crash-safe tra…
Cowork Bench
Multi-tool LLM agent benchmark: 496 end-to-end MCP tasks, deterministic evaluation, Russian localization
Openbb Workspace Bench
Benchmark for agents operating OpenBB Workspace through MCP
Unreal Agent Benchmark
Evidence-based Unreal Engine 5.8 benchmark for complete AI game-building systems using Epic Unreal MCP, playable tasks, packaging, and launch proof.
Adv Bench
MCP Adversarial Benchmark; MCP利用時の敵対的入力防御を測定するためのベンチマーク
Oh No My Claudecode
The coding harness that measures itself. Wraps Claude Code / Codex / OpenCode, blocks false "done" with an independent verifier, mediates side effects, writes tamper-evident recei…
Dew Flow Benchmark
Measure any code repository, at any commit, through any retrieval engine — with the sample sizes attached.
Zcode Autoresearch
ZCode 自主实验循环插件，让 coding agent 基于固定度量自动迭代优化
CodeContextBench Dashboard
Benchmark platform for measuring how code-context tools (Sourcegraph MCP) affect coding-agent performance: adapter generation, local validation, GCP execution, and a Streamlit res…
Vmbench
Cross-platform VPS benchmark CLI: sysbench/fio/openssl hardware tests, network diagnostics (route/ping/speed/IP quality), TUI, JSON/HTML reports, MCP server. YABS alternative.
Humanloopbench
Human-in-the-loop LLM Failure Detection & Benchmark Toolkit — Auto-flag LLM failures (hallucination, refusal, factual error) with LLM-as-Judge explanations. Confirm them in a revi…
Tool Output Spoofing Lab
Benchmarking schema-valid false tool observations and defense baselines for tool-using LLM agents.
Nuxs
Up to 99% token savings on AI agent context. Free. Provider-agnostic (Claude, Cursor, Codex, Cline, any MCP agent). 91.62% over 626.8M auditable tokens. Try: nuxs.ai/playground
Agent Governance Toolkit
Manage and monitor AI agents in production environments with tools for governance, safety, and compliance.
Duke Gta
Duke Graduate TA Engineering Portfolio: AI agents, MCP, Debugger Benchmark, Docker, CI/CD, WASM
Token Benchmark
Token-cost benchmark for MCP tool definitions: 9 servers measured, a 25x efficiency spread, and a 95% lighter redesign of the worst offender
Aobench
Role-aware, permission-enforced benchmark for LLM agents that operate HPC systems — SLURM, telemetry, RBAC, docs, facility. A policy violation hard-fails the task. 88 tasks (10 ca…
Motion Dev MCP Poc
A proof-of-concept benchmark evaluating AI coding agents (Claude Code, Codex, Gemini CLI) building Motion.dev animations.
Coordination Noise Floor Protocol
Code, data & paper: How Much Coordination Gain Is Real? A Paired Noise-Floor Protocol for Multi-Agent LLM Benchmarks (KDD 2026 Workshop)
Agentbench
Automated benchmarking framework for evaluating AI agent security
Agent Memory Torture Test
Memory Integrity Bench: can your agent remember the truth after the truth changes?
Qector Decoder Workbench Windows
QECTOR Decoder Workbench v1.0.7 (Windows) — 19 decoders, 10 code families, 85-tool MCP Server (qec.* 3-6 tools scoped), versioned schemas, immutable manifests with git hash + pari…
Betterthantoon
GCF vs TOON benchmark data
DB Benchmark
Benchmarking PostgreSQL, MySQL, SQL Server, and SQLite under AI agent MCP workloads
Tournament
Benchmark LLMs on your own scenarios with a multi-judge panel. Full MCP server (tools, resources, prompts, progress) plus local GUI, CLI, and desktop app. One OpenRouter key.
Ecosystem Measurements
Reproducible measurements of the MCP registry, the VS Code gallery, Glama, hosted MCP server endpoints and GitHub search. Twelve findings, each with the script, its raw dated outp…
Special War
A jeopardy-style CTF where the competitors are AIs. Play via MCP, no account needed.
Benchmark Article Resources
An Investigation Across Two MCP Servers, Two Frontier Models, and 20 Controlled Benchmark Runs
Gtm Arena
The open benchmark for AI marketing agents. Blind pairwise comparisons with ELO rankings for LinkedIn posts, cold emails, and ad copy.
Local Five Agent System
Local Ollama five-agent system with Gemma/Qwen roles, benchmark harness, MCP memory capture, and WinUI workbench.
A2B
[L3 ASSET] A2B — AssetOpsBench Bridge. IJCAI-25. Constitutional governance benchmarks for industrial AI.
Assayer Memory MCP
A local knowledge base your agents can read — and the engines that keep it honest. Computed themes, measured links, budgeted context over MCP, a verdict funnel that drains its own…
Ollamancer
Fully-local terminal AI agent for Ollama. 34 tools, MCP, local RAG, deterministic hallucination checks. No cloud, no API keys, nothing leaves your machine.
Ukrainian LLM Eval
Reproducible Ukrainian-language evaluation of language models and agents, with optional reference tools.
Storm
Fault-injection proxy at the MCP boundary: what tool failures cost your agent (completion, token multiplier, retry storms, dollars per task). Preregistered dose-response.
Nova Rag Benchmark
Benchmark for Code RAG MCP Servers — measure how well RAG helps AI find the right code. Hit@K, MRR, latency, A/B vs grep baseline, leaderboard.
LLM Eval
MCP server for LLM evaluation gates.
Benchmark Search Tool MCP
Claude Code 搜索工具对比评测：5 工具 × 15 测试用例 × 52 组合，LLM-as-Judge 验证
Wikibench
wikipedia racing for LLMs!
Damn Vulnerable Agent Asset Corpus
Damn Vulnerable Agent Asset Corpus: benchmark fixtures for agent-asset scanners and AAC release assurance.
Gemma4 Bench
MacBook Air M2 24GB 한 대로 돌리는 Claude × 로컬 LLM 하이브리드 멀티-티어 오케스트레이션 — 컨셉·벤치마크·실전 셋업
LLM Framework Benchmark
Can an AI coding agent build the same framework-neutral spec on different stacks? External-oracle-graded benchmark. First: Spring Boot vs Tiko DI.
Vulnbench
Benchmark of static analyzers on real, publicly disclosed vulnerabilities in open-source MCP servers: vulnerable/fixed commit pairs, Docker harness, SARIF scoring
Divergence
Precision-first security analyzer for MCP servers and agent skills. 0% false positives on traps, 96% recall.
Image Servers Tested
21 image MCP servers, same 8 prompts, what came back and the transcripts
Scoreia Forge Results
Open results of La Forge (ScoreIA): AI models forge and animate a 3D knight over MCP, scored by a programmatic referee (no LLM judge).