Stats Explorer runs on your own machine — the client starts it, so there is no endpoint to ping. 295 installs a week from npm. Last commit 24 Oct 2026.
Verified statistics for AI agents: guided tests, A/B tests, Tukey HSD, reports, optional Jev checks.
Today is the operative word: we check Stats Explorer every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add stats-explorer-mcp -- npx -y stats-explorer-mcp
{
"mcpServers": {
"stats-explorer-mcp": {
"command": "npx",
"args": [
"-y",
"stats-explorer-mcp"
]
}
}
}
[mcp_servers.stats-explorer-mcp]
command = "npx"
args = ["-y", "stats-explorer-mcp"]
{
"mcpServers": {
"stats-explorer-mcp": {
"command": "npx",
"args": [
"-y",
"stats-explorer-mcp"
]
}
}
}
{
"servers": {
"stats-explorer-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"stats-explorer-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
STATS_EXPLORER_ROOTS (Folders the tools may read data files from and write reports to (separated by : on macOS/Linux, ; on Windows). Defaults to the client's workspace roots, or the current directory), STATS_EXPLORER_STREAM_MB (CSV/TSV/Parquet files above this size (MB) are streamed in parallel passes; any size works. Default 512), STATS_EXPLORER_WORKERS (Parallel readers for large files. Default: all CPU cores but one (max 16)), STATS_EXPLORER_CACHE (Keep columns read from large files for fast follow-up questions: ask (default), on or off), STATS_EXPLORER_CACHE_DIR (Folder for saved large-file columns. Default: the system temp folder), STATS_EXPLORER_MAX_FILE_MB (Refuse data files larger than this (MB). Default: no limit), STATS_EXPLORER_GUIDED (on (default): classify each request and follow the six-step workflow for hypothesis tests. off: expose the individual test tools directly), STATS_EXPLORER_CONFIRM_SELECTIONS (on (default): the user confirms every column, chart type and report section before anything runs. off: run with the AI's choices), STATS_EXPLORER_CONFIRM_VIA (auto (default): forms where the client supports them. chat: always ask in the conversation (recommended for Claude Code)), STATS_EXPLORER_HIDE_LINK (Set to 1 to remove the one-line Stats Explorer app link from results), STATS_EXPLORER_JEV (Independent checks by Jev (TypeSafe): off (default), cooperate (the AI calls your Jev MCP server), enforce (Stats Explorer calls the Jev API) or orchestrated (signed decisions from an orchestrator)), STATS_EXPLORER_JEV_CHECKPOINTS (Comma list of plan, conclusions, data. Default plan,conclusions (data sends column names and text samples to Jev)), TYPESAFE_API_KEY (Enforce mode only: TypeSafe API key Stats Explorer uses to call Jev), STATS_EXPLORER_APPROVAL_SECRET (Orchestrated mode only: shared secret (32+ characters) the orchestrator signs decisions with), STATS_EXPLORER_AUDIT_LOG (Path of a JSON Lines file recording every Jev checkpoint decision).
Paid x402 facts for AI agents: signed statements, door reports, endpoint checks, chain reads
Local-first, read-only ECZ-ID verifier over MCP (stdio): checks, reports, explains, routes.
Calibrated vision for AI agents: meal nutrition, personal color, styling reports. x402, no API key.
Verified agentic-commerce protocol status + merchant agent-readiness reports. Free on testnet.
Score repos for launch and agent readiness, find verified agents, read collaborations. Read-only.
QA platform for agents: coverage signals, in-repo test plans, verified tests and release governance.
Verifies AI agent work end to end: real artifacts and outcomes checked, not self-reported success.
MCP server for fixing static analysis issues from SARIF reports with AI agents
Answers built from our own checks of this server.