GPTProto MCP
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GPTPROTO_API_KEY | No | GPTProto API key. If not set, the server uses the key stored by the GPTProto CLI in ~/.config/gptproto/key. | |
| GPTPROTO_API_BASE_URL | No | Optional base URL for the GPTProto API. Defaults to the value stored by the CLI (or the official API). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| gptproto_statusA | Check whether the MCP uses the default API origin, whether an API key exists, and whether file uploads are enabled. Never returns the key or a custom origin. |
| gptproto_models_listA | List live GPTProto models, optionally filtered by one broad capability. Use before choosing a model. |
| gptproto_model_describeA | Return a model's live methods, paths, native parameters, enums, response type, and asynchronous polling contract. |
| gptproto_pricingA | Read GPTProto's live public price catalog. Query one exact provider/model, search all models, or filter a capability and sort by its starting price. |
| gptproto_requestA | MCP equivalent of |
| gptproto_custom_createC | Submit an existing GPTProto custom image, video, audio, lip-sync, 3D, or image-edit request using its documented native body. |
| gptproto_tasks_listA | List recent asynchronous tasks saved locally, including their IDs, model, type, time, state, and completed result URLs. Use this after an interrupted tool call or in a new conversation before creating another task. |
| gptproto_task_getA | Perform exactly one status query for a saved task after an interrupted automatic wait. It never resubmits or internally loops. |
| gptproto_task_waitA | Resume MCP-owned polling for an existing known task after an interrupted creation call. It never resubmits and is capped at 600 seconds. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Most tools are clearly distinct, but task_get and task_wait both handle interrupted operations and could be confused at a glance. custom_create and request also overlap in submitting native API bodies, though descriptions clarify the specific use case.
All tools share the gptproto_ prefix, which is consistent, but the remaining pattern mixes noun-only names (status, pricing, request) with noun_action names (models_list, model_describe, task_get). Singular/plural usage is inconsistent (models_list vs task_get vs tasks_list).
Nine tools is well-scoped for an API client covering status, models, pricing, request execution, and task management. Each tool has a clear role and none feel redundant.
The tool surface covers the core lifecycle of discovering models, pricing, making requests, and managing async tasks. Minor gaps exist, such as no explicit task cancellation or file upload tool, despite status checking upload capability.