Skip to contentLLM Eval Harness MCP Server | MCPpediaTrust scoreTrust score 52 of 100, Fair Fair trust
52 of 100 from six public signals. Nobody can buy a point.
- Listed in the official MCP registry.
- No license recorded.
How we score trust Trust breakdown
Where the 52 comes from
22 + 20 + 0 + 0 + 7 + 3 = 52. Each factor is capped, so no single signal can carry a listing.
Registry22 / 30
Active in the official MCP registry, plus a known vendor namespace.
Registry: 22 of 30 points.Maintenance20 / 20
How recently the repository saw a commit or release.
Maintenance: 20 of 20 points.Popularity0 / 20
GitHub stars and weekly downloads, on a log scale.
Popularity: 0 of 20 points.License0 / 10
A clear, OSI-style open-source license.
License: 0 of 10 points.Security7 / 10
Not archived, a secure install path, a pinned version.
Install
Pick your client, copy the config, paste it in.
~/Library/Application Support/Claude/claude_desktop_config.json
{ "mcpServers": { "razilzul-llm-eval-harness": { "command": "docker", "args": [ "run", "-i", "--rm", "ghcr.io/razilzul/llm-eval-harness:1.0.0" ], "env": { "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY", "GEMINI_API_KEY": "YOUR_GEMINI_API_KEY" } } }}
Security: 7 of 10 points.
Docs3 / 10
A real summary, a README with depth, listed tools.
Docs: 3 of 10 points.Environment variables
| Name | Required | Notes |
|---|
ANTHROPIC_API_KEYSecret | Optional | Only needed to evaluate anthropic:* models. Not needed for mock models. |
GEMINI_API_KEYSecret | Optional | Only needed to evaluate gemini:* models. Not needed for mock models. |
Tools show up when the registry metadata or the repository README lists them. GitHub enrichment fills this in when a token is configured.
About
Run a golden set and gate LLM model swaps on accuracy, with latency and cost per request.
FAQ
- ?What is LLM Eval Harness?
- Run a golden set and gate LLM model swaps on accuracy, with latency and cost per request.
- ?How do I install LLM Eval Harness?
- Pick a client above and copy the generated config. Cursor and VS Code also have a deep link when the server can run locally or remotely.
- ?Is LLM Eval Harness official?
- No. Registry-listed is not the same as vendor-official. Read the methodology for the difference.
- ?How is the trust score calculated?
- This listing scores 52 (Fair). The formula is published on the methodology page.
Reviews
No average · 0 publishedNo reviews yet. Used it? Tell the next person how it went.
Sign into review this server.
MM74Trust score 74 of 100, Good
Hosted agent memory that learns from outcomes, with shared rooms. No key: mcp.mnemoverse.com/mcp
27 stars290/wk weekly downloads— tools
Op72Trust score 72 of 100, Good
Investment runtime for AI agents — LLM investment committee, portfolio ledger, decision accounting
114 stars—/wk weekly downloads— tools
Kn72Trust score 72 of 100, Good
Persistent memory for AI agents, for code, research and everyday life. Stale facts retire.
105 stars—/wk weekly downloads— tools
MM70Trust score 70 of 100, Good
Local-first, cloud-friendly memory — 100+ tools, 99.2% LongMemEval-S retrieval@8, hybrid search.
29 stars—/wk weekly downloads— tools