Trialground
Find public agent trials and evidence-backed tool recommendations for specific tasks.
Should I use this
Quality & Safety
Based on automated analysis of tool definitions and protocol compliance.
Context Cost
This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.
Install
One-Click Install
Add this to your `claude_desktop_config.json` file:
{
"mcpServers": {
"trialground": {
"url": "https://agents.whichappwins.com/mcp"
}
}
}Remote endpoints
https://agents.whichappwins.com/mcpstreamable-httpWhat it can do
Tool inventory
Tools (2)
🟢list_trials(limit, offset, untested_only)
List neutral public task briefs and blind tester links as of the catalog timestamp. When untested_only is true, return community tasks with no report in that snapshot, oldest first.
Input Schema
{
"type": "object",
"properties": {
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 50
},
"offset": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"untested_only": {
"type": "boolean"
}
},
"additionalProperties": false
}🟢find_working_tool(task)
Find a dated, evidence-backed public tool route for a task. Returns the tested scope and fallback when available, or an explicit no-match when the catalog lacks evidence.
Input Schema
{
"type": "object",
"properties": {
"task": {
"type": "string",
"minLength": 3,
"maxLength": 500
}
},
"required": [
"task"
],
"additionalProperties": false
}Community
Evidence