MCPMark
Stress test for realistic MCP use across Notion, GitHub, filesystem, PostgreSQL and Playwright servers.
Reference by EVAL SYS, with LobeHub and NUS TRAIL
IndexedApache-2.0★ 467
About
Whether an agent can complete multi-step create, read, update and delete workflows through MCP servers, checked by scripts.
Use cases: llm-evaluation
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
About the benchmark
- Benchmark site
- mcpmark.ai/
- Maintainer
- EVAL SYS, with LobeHub and NUS TRAILmcpmark.ai
- Launched
- 2025-09arxiv.org
- Leaderboard
- mcpmark.ai/leaderboard/verified
We link to the official leaderboard instead of copying scores. Benchmarks get gamed and saturate; we never use them to rank listings.
Signals
- Stars
- 467
- Forks
- 50
- Open issues
- 19
- Last commit
- Jun 12, 2026 (4 months ago)
- Last release
- Sep 20, 2025 (1 year ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/eval-sys/mcpmark
- Website
- mcpmark.ai/
- Docs
- mcpmark.ai/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
Cite this page
MCPMark benchmark. Agent Mart. https://agentmart.dev/reference/mcpmark (last updated 2026-10-24)