QVeris — Claude Code MCP server: what it does, how to install
Enables AI agents to find and execute external tools through natural language queries. Includes parameter validation and credit auditing.
claude mcp add --transport http mcp https://mcp.qveris.ai/mcpWhat QVeris does
The server provides an interface for discovering and running external capabilities on demand. Instead of configuring separate connections for every service, the model queries discover in plain language and selects candidate tools.
For verification, the server exposes inspect and probe. The probe step checks parameter schemas before execution. It generates zero-cost estimates without consuming balance. Actual execution runs through call with a preserved search_id. Using the server requires an active QVERIS_API_KEY.
What it helps with
- discover relevant third-party tools using natural language prompts
- validate input schemas and obtain cost quotes before running tools
- execute external capabilities through a single unified gateway
- inspect tool contracts and audit credit consumption safely
Example prompts for the agent
- Find a weather tool and get the current forecast for Tokyo
- Run probe on the candidate tool to validate parameters without spending credits
- Execute the discovered email service using the returned search_id
How to install QVeris
1Platform CLI: remote server
claude mcp add --transport http mcp https://mcp.qveris.ai/mcp2Platform CLI
claude mcp add --env QVERIS_API_KEY=<QVERIS_API_KEY> --env QVERIS_BASE_URL=<QVERIS_BASE_URL> --transport stdio mcp -- npx -y @qverisai/mcpReplace <…> with your own values
1Platform CLI: remote server
codex mcp add mcp --url https://mcp.qveris.ai/mcp2Platform CLI
codex mcp add mcp --env QVERIS_API_KEY=<QVERIS_API_KEY> --env QVERIS_BASE_URL=<QVERIS_BASE_URL> -- npx -y @qverisai/mcpReplace <…> with your own values
Config file
{
"mcpServers": {
"mcp": {
"url": "https://mcp.qveris.ai/mcp"
}
}
}{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@qverisai/mcp"
],
"env": {
"QVERIS_API_KEY": "<QVERIS_API_KEY>",
"QVERIS_BASE_URL": "<QVERIS_BASE_URL>"
}
}
}
}Replace <…> with your own values
Config file
{
"servers": {
"mcp": {
"type": "http",
"url": "https://mcp.qveris.ai/mcp"
}
}
}{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@qverisai/mcp"
],
"env": {
"QVERIS_API_KEY": "<QVERIS_API_KEY>",
"QVERIS_BASE_URL": "<QVERIS_BASE_URL>"
}
}
}
}Replace <…> with your own values
1Platform CLI: remote server
gemini mcp add -t http mcp https://mcp.qveris.ai/mcp2Platform CLI
gemini mcp add -e QVERIS_API_KEY=<QVERIS_API_KEY> -e QVERIS_BASE_URL=<QVERIS_BASE_URL> mcp npx -y @qverisai/mcpReplace <…> with your own values
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about QVeris
How to install QVeris in Claude Code?
Run claude mcp add --transport http mcp https://mcp.qveris.ai/mcp. Alternatively, configure stdio transport with claude mcp add --env QVERIS_API_KEY=<QVERIS_API_KEY> --env QVERIS_BASE_URL=<QVERIS_BASE_URL> --transport stdio mcp -- npx -y @qverisai/mcp.
What does QVeris do?
It provides dynamic tool discovery and execution for LLMs through the QVeris API, including parameter validation, cost probing, and usage auditing.
Is running probe mandatory before calling a tool?
No, probe is optional. It is intended for validating schema constraints or verifying cost estimates before committing an actual execution via call.
