MCP servers and skills: search, filter by where they run, and compare.
The home for your AI agents. Live dashboard, public profiles, README badges, LLM eval, and failure insights — one line of code.
Hosted by authorMITFew starsNo key
Run `doca_bench` (DOCA 2.7.0 or newer) to measure throughput, bulk latency, precision latency, or maximum bandwidth for RDMA, Compress, AES-GCM, SHA, DMA, EC, Ethernet, Comch, or GPUNetIO on a host or
OfficialSkillApache-2.0★ 3.6k in collectionNo key
AI-assisted outbound sales pipeline — FastAPI backend exposing the same service layer over HTTP and MCP
MCPLink onlyFew stars
Claude Code plugin that generates ADE-Bench benchmark tasks from your own dbt project
SkillMIT★ 3Few starsNo key
博主的对标复盘雷达：每周导入自己和对标账号的数据，标出异常爆款、生成数据台账和周报
SkillApache-2.0Few starsNo key
Score Python MCP defences on mcp-defense-bench: same metrics as the JS leaderboard, plus a pytest regression gate.
PythonApache-2.0Few starsNo key
A test bench for tuning Grok CLI generation settings. Dashboard, rewriting proxy, system-prompt section editor.
SkillMIT★ 4Few starsNo key
Reliability by construction for long-horizon healthcare agents on χ-Bench (IEEE Big Data Cup 2026, Challenge 07): contract-checklist hooks, critic panel and closure gate on the official Claude Code ha
SkillApache-2.0★ 1Few starsNo key
Open, reproducible benchmarks for crypto infrastructure: aggregators, bridges, blockchains, perps. Live data, public methodology.
Hosted by authorMIT★ 8No key
A frontier benchmark for enterprise-scale data science
SkillApache-2.0★ 8No key
Provide an open test corpus to assess and improve AI agent egress security through validated cases and automated workflows.
SkillApache-2.0★ 1Few starsNo key
Free US drug-price benchmarks from federal data (CMS NADAC): a benchmark, not a price anyone owes.
Hosted by authorNo licenseFew starsNo key
The code to visualize colon-bench (MICCAI 2026) data and run evaluations of MLLMs on the benchmark.
SkillLink only★ 5
Build benchmarks from your own account history, so you can tell a real change from normal week-to-week noise.
SkillMITFew starsNo key
Berkeley Agentic AI class: Enhanced Finance Agent Benchmark with AgentBeats
MCPLink only★ 2Few stars
Claude skill: router for subscription benchmark questions that picks the right report and never mixes metrics with different denominators.
SkillLink onlyFew stars
Benchmark harness: CLI vs MCP (eager/deferred) vs Code Mode. 5 tasks × 4 configs × 10 runs = 200 data points. Token counts from Anthropic API usage object.
MCPLink onlyFew stars
Pink Agentic AI Payments: open benchmark of what Claude and Gemini agents try when asked to get around a spending rule (80 runs, Oct 2026). Data CC BY 4.0, code MIT.
MCPLink onlyFew stars
Real-time multimodal VLM-to-action inference pipeline + multi-framework benchmark harness (NIM Qwen3-VL, vLLM, SGLang, TensorRT-LLM, ModelOpt, Triton) targeting RTX 5090, RTX PRO 6000 Blackwell, and H
SkillLink only★ 2Few stars
UX Score and Web Vitals (TTFB, FCP, LCP, INP, CLS) for any website, from real-user CrUX data.
Hosted by authorNo licenseFew starsNo key
The luck-robust benchmark for trading agents.
Node.jsApache-2.0★ 6No key
A standalone benchmark-store MCP. It ingests code-size and on-target cycle measurements into a provenance-bound SQLite store and exposes Pareto / CI analytics.
PythonMITFew starsNo key
Open daily datasets on the AI economy: model prices across providers, price changes, company figures with sources, Mindshare, GPU rental prices, Gecko Tests. CC BY 4.0.
MCPLink onlyFew stars
aicharts plots published AI benchmark scores against cost and tokens per task, and a local collector measures your own agents' token use.
SkillMIT★ 4Few starsNo key
Who stands behind it
