model-GPT

Sends prompts to the OpenAI API with automatic model selection and token logging.

1|Updated May 24, 2026

Search Tags:#api integration #reasoning #model routing #openai #gpt #prompting #token logging

Ready to run

npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill model-gpt

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: model-GPT
Source: https://github.com/tinner-deinno/innova-skills-lib/tree/main/core/model-GPT
Command: npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill model-gpt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The model-GPT skill eliminates the time spent manually picking and calling the correct OpenAI model by routing your task to an appropriate GPT/o1/o3 variant and returning the response with consistent handling.

Core Features & Use Cases

Direct OpenAI API execution: Sends your prompt straight to OpenAI using your configured OPENAI_API_KEY for reliable general availability.
Model selection with auto-routing: Chooses models based on task keywords such as reasoning, quick summaries, and long-context needs.
Operational commands: Supports /model-GPT [prompt], --model for explicit selection, --list to view models, and --status to test connectivity.
Usage observability: Logs token usage to ~/Jit/ψ/telemetry/token_log.jsonl for tracking and auditing.

Quick Start

Run /model-GPT Explain how multi-agent systems work in Thai to get a best-fit OpenAI GPT response using the default model.

Frequently Asked Questions about model-GPT

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automatically route prompts to the right OpenAI model in an agent workflow?▼

Automatic model routing sends your prompt to an appropriate GPT or reasoning variant by analyzing task keywords like long-context or deep reasoning, returning completions without manual selection.

How do I use an OpenAI API key for generating task-specific completions?▼

To generate task-specific completions, configure your OPENAI_API_KEY in the ~/Jit/.env file, then execute your prompt command to send requests directly to the OpenAI API for processing.

Can I explicitly select an OpenAI GPT model instead of using auto-routing?▼

Yes, you can bypass auto-routing logic by passing the --model flag with your prompt command to explicitly select a specific OpenAI GPT or reasoning model for the task.

How do I log token usage for OpenAI API calls in my agent workflow?▼

Token usage logging happens automatically during API execution, recording token counts and usage data to ~/Jit/ψ/telemetry/token_log.jsonl to enable tracking and auditing of OpenAI API consumption.

Does this OpenAI API integration support long-context and deep reasoning tasks?▼

Yes, the integration supports long-context and deep reasoning tasks by routing prompts to appropriate OpenAI reasoning models, alongside handling general Q&A, summarization, formatting, and translation.

What are the limitations of using auto-routing for OpenAI model selection?▼

Auto-routing relies on task keywords to select models, which may not always match complex or ambiguous prompt requirements; use --model for explicit selection when precise control over the OpenAI variant is needed.