Run a golden set and gate LLM model swaps on accuracy, with latency and cost per request.
Connect Runs on your machine
claude mcp add llm-eval-harness -e ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY> -e GEMINI_API_KEY=<GEMINI_API_KEY> -- docker run -i --rm ghcr.io/razilzul/llm-eval-harness:1.0.0
{
"mcpServers": {
"llm-eval-harness": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/razilzul/llm-eval-harness:1.0.0"
],
"env": {
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}
{
"mcpServers": {
"llm-eval-harness": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/razilzul/llm-eval-harness:1.0.0"
],
"env": {
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}
{
"servers": {
"llm-eval-harness": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/razilzul/llm-eval-harness:1.0.0"
],
"env": {
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}
Replace the values in angle brackets with your own. Keys stay on your machine; we never see them.
Tools
This server runs locally, so its tools are listed by your MCP client once it starts.
Environment variables
| Name | Description |
|---|---|
ANTHROPIC_API_KEYOptional, secret | Only needed to evaluate anthropic:* models. Not needed for mock models. |
GEMINI_API_KEYOptional, secret | Only needed to evaluate gemini:* models. Not needed for mock models. |
Details
| Package | ghcr.io/razilzul/llm-eval-harness:1.0.0 (OCI, stdio) |
| First published | 18 Oct 2026 |
| Last updated | 18 Oct 2026 |
| Registry name | io.github.razilzul/llm-eval-harness |
More ai and media servers See all
Listing data from the official MCP registry. Status and tools from NestDaddy's daily check. Browse all MCP servers.
