Browse
→ AI & Models
→ GPTProto MCP
GPTProto MCP
TypeScript stdio MCP server that connects MCP hosts to GPTProto public APIs for model discovery, live pricing, native API requests, custom generation, and asynchronous task polling. It reads credentials from GPTProto CLI configuration files or GPTPROTO_API_KEY, does not embed a model catalog, and exposes tools for status, model listing, model description, pricing lookup, direct documented requests, and custom task creation/polling. Public GitHub repository created September 20, 2026.
MCP unverified
Integration
| Transport | stdio |
| Auth | api-key |
| Endpoint | npx --yes --package=github:gptproto-ai/gptproto-mcp gptproto-mcp |
| Install | npx --yes --package=github:gptproto-ai/gptproto-mcp gptproto-mcp |
Use Cases
| 01 | Ask an assistant to compare GPTProto live model capabilities and pricing before choosing a route |
| 02 | Call documented GPTProto-compatible routes from a local MCP host without exposing the API key in chat |
| 03 | Create and poll asynchronous GPTProto tasks through MCP tools instead of a custom client |
Tags
gptproto model-discovery pricing generation async-tasks typescript node stdio
Machine-readable: /api/servers.json
· JSON-LD schema embedded in <head>