Prerequisites
- Node.js 18 or later
- An AskVedicGuru developer API key (see Authentication)
- An MCP-compatible AI host
Install
npm install -g @askvedicguru/mcpOr use without a global install via npx:
npx @askvedicguru/mcpConfigure your API key
export ASKVEDICGURU_API_KEY=avg_xxxxxxxxxxxxxxxxxxxxOr pass it directly when testing:
ASKVEDICGURU_API_KEY=avg_... npx @askvedicguru/mcpAI Host Configurations
Each host below uses npx for zero-install convenience. Replace with your globally installed command (@askvedicguru/mcp) if preferred.
Claude Desktop
Open ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows) and add:
{
"mcpServers": {
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}Restart Claude Desktop. The AskVedicGuru tools will appear in the tools panel.
Claude Code
Add to ~/.claude/settings.json (or your project's .claude/settings.json):
{
"mcpServers": {
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}Then restart Claude Code and ask: "List my saved profiles"
Cursor
Open Settings → MCP Servers in Cursor, then add:
{
"mcpServers": {
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}Or manually edit ~/.cursor/mcp.json.
Windsurf (Codeium Cascade)
Open Windsurf → Settings → MCP, then add:
{
"mcpServers": {
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}Roo Code (formerly Cline)
Open Roo Code settings → MCP Servers and add:
{
"mcpServers": {
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}Roo Code will now offer AskVedicGuru tools during task execution.
Continue.dev
Add to .continue/config.ts in your project:
export function modifyConfig(config: Config): Config {
config.mcpServers ??= [];
config.mcpServers.push({
name: "askvedicguru",
command: "npx",
args: ["-y", "@askvedicguru/mcp"],
env: {
ASKVEDICGURU_API_KEY: "avg_xxxxxxxxxxxxxxxxxxxx",
},
});
return config;
}Cherry Studio
Open Settings → MCP → Add Server and fill in:
| Field | Value |
|---|---|
| Name | askvedicguru |
| Command | npx |
| Args | -y @askvedicguru/mcp |
| Environment | ASKVEDICGURU_API_KEY=avg_xxxxxxxxxxxxxxxxxxxx |
ChatWise
Open Settings → MCP Servers → Add and paste:
{
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}Zed
Add to your project's .zed/settings.json:
{
"context_servers": {
"askvedicguru": {
"command": {
"path": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}
}Bolt.new
Bolt.new supports MCP via configuration. Add to your project root mcp.json:
{
"mcpServers": {
"askvedicguru": {
"command": "npx",
"args": ["-y", "@askvedicguru/mcp"],
"env": {
"ASKVEDICGURU_API_KEY": "avg_xxxxxxxxxxxxxxxxxxxx"
}
}
}
}Environment variables
| Variable | Description |
|---|---|
ASKVEDICGURU_API_KEY | Required. Your developer API key |
ASKVEDICGURU_API_URL | Optional. Custom API base URL (self-hosted) |
Verify
Ask your AI assistant:
"List my AskVedicGuru profiles"
The assistant should call list_profiles and return your saved profiles.