Read-only Model Context Protocol (MCP) server for reusable AI prompts and assistant blueprints. Python, local-first, no write access.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add ai-workbench -- uvx alptugharun-ai-workbench-mcp~/.codex/config.toml
[mcp_servers.ai-workbench]
command = "uvx"
args = ["alptugharun-ai-workbench-mcp"]
# codex mcp add ai-workbench -- uvx alptugharun-ai-workbench-mcp~/.cursor/mcp.json
{
"mcpServers": {
"ai-workbench": {
"command": "uvx",
"args": [
"alptugharun-ai-workbench-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"ai-workbench": {
"type": "stdio",
"command": "uvx",
"args": [
"alptugharun-ai-workbench-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"ai-workbench": {
"command": "uvx",
"args": [
"alptugharun-ai-workbench-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"ai-workbench": {
"command": "uvx",
"args": [
"alptugharun-ai-workbench-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"ai-workbench": {
"command": "uvx",
"args": [
"alptugharun-ai-workbench-mcp"
]
}
}
}