Security & Testing
polygraph.so
Provides tools to check, list, and request behavioral grades for servers from a grading service.
ENDPOINT 1
https://polygraph.so/api/mcp
MCP server metadata
- Name
- polygraph
- Version
- 1.0.0
polygraph publishes independent behavioral security grades (A-F) for MCP servers. Use check_server as the pre-flight check before recommending or installing a server: it returns the published grade in well under a second and runs nothing. A not_available result means the server is unevaluated (neither safe nor unsafe), not that it failed. Call request_grade to get it graded: a $1 one-time fee applies, paid via the response's payment link (x402 for agents, web checkout for humans); the fee buys the run, never the grade. list_servers returns servers with a published grade, 25 at a time by default (up to 100 per call), with a summary that always covers the full graded corpus. Every grade is reproducible: the report page carries a one-command re-run. This endpoint does not grade servers itself.
Known tools 3
CONNECT WITH APPROVAL
Client installation
Review this server and its permissions before adding it. Secret placeholders must be set locally.
Codex
~/.codex/config.toml
[mcp_servers.polygraph]
url = "https://polygraph.so/api/mcp"
enabled = true
Claude Code
.mcp.json
{
"mcpServers": {
"polygraph": {
"type": "http",
"url": "https://polygraph.so/api/mcp"
}
}
}
Claude Desktop
Settings → Connectors → Add custom connector
Name: polygraph
Remote MCP URL: https://polygraph.so/api/mcp
Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.
Cursor
.cursor/mcp.json
{
"mcpServers": {
"polygraph": {
"url": "https://polygraph.so/api/mcp"
}
}
}
Visual Studio Code
.vscode/mcp.json
Add to Visual Studio Code{
"servers": {
"polygraph": {
"type": "http",
"url": "https://polygraph.so/api/mcp"
}
}
}
Generic MCP
Client-specific MCP configuration
{
"name": "polygraph",
"transport": "streamable-http",
"url": "https://polygraph.so/api/mcp"
}
MCP Inspector
Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.
TRUST AND VERIFICATION EVIDENCE
Loading Trust v2 evidence…
Checking the associated registrable domain. The BuiltWith key remains server-side.
Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.