MCP Proof: Bounded MCP Evaluation
DeprecatedPricing
Pay per usage
Run a bounded known-answer evaluation set against an MCP server, snapshot tool schemas, grade deterministically, and return an evidence receipt. Not certification.
You can access the MCP Proof: Bounded MCP Evaluation programmatically from your own applications by using the Apify API. You can also choose the language preference from below. To use the Apify API, you’ll need an Apify account and your API token, found in API & Integrations in Apify Console.
$echo '{< "url": "https://mcp.example.com/mcp",< "evaluationSet": {< "schema": "mcp-proof/evaluation-set/1",< "name": "my-smoke-set",< "version": "1.0.0",< "expectedTools": [< "lookup_zip"< ],< "cases": [< {< "id": "zip-97201-portland",< "tool": "lookup_zip",< "arguments": {< "zip": "97201"< },< "expect": {< "kind": "textContainsAll",< "values": [< "Portland"< ]< }< }< ]< }<}' |<apify call savagnic/mcp-proof --silent --output-datasetThe Apify CLI is the official tool that allows you to use MCP Proof: Bounded MCP Evaluation locally, providing convenience functions and automatic retries on errors.
Using installation script (macOS/Linux):
$curl -fsSL https://apify.com/install-cli.sh | bashUsing installation script (Windows):
$irm https://apify.com/install-cli.ps1 | iexUsing Homebrew:
$brew install apify-cliUsing npm:
$npm install -g apify-cliOther API clients include: