skills.rest
tarabakz25
tarabakz25COMMUNITY
Explore all skills by this author →
Website𝕏@kz25_kmcGitHub

openai-docs

Retrieve official OpenAI documentation and GPT-5.4 upgrade guidance via MCP tools.

1|Updated Jul 2, 2026
Search Tags:#documentation#api#mcp#openai#gpt-5.4#model migration#prompt upgrade
One-click install
npx skills add https://github.com/tarabakz25/my-skills --skill openai-docs-tarabakz25
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openai-docs
Source: https://github.com/tarabakz25/my-skills/tree/main/.system/openai-docs
Command: npx skills add https://github.com/tarabakz25/my-skills --skill openai-docs-tarabakz25

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Developers building with OpenAI APIs often rely on outdated or unofficial information when choosing models, migrating to newer versions, or updating prompts. This Skill grounds every answer in current official OpenAI documentation with citations, and provides a disciplined workflow for model and prompt upgrades. ## Core Features & Use Cases - Authoritative docs lookup: Searches and fetches pages from the developers.openai.com MCP server, falling back to official-domain web search only when MCP returns nothing useful. - Model selection guidance: Resolves the latest recommended model for a use case using the live latest-model guide, with bundled references as fallback. - Model and prompt upgrades: Runs a resolver script to detect the current model target, then applies a narrow upgrade workflow that classifies changes as model-string-only, light prompt rewrite, or blocked. - Use Case: A user asks to migrate their app from gpt-4o to the latest model. The Skill fetches current migration guidance, inventories model usage, and recommends a minimal, behavior-preserving upgrade plan with reasoning-effort settings. ## Quick Start Ask the agent to look up the latest OpenAI model for your use case or to plan an upgrade of your existing integration to the newest GPT model with cited official guidance.

Frequently Asked Questions about openai-docs

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

FAQPage Schema
How do I find the latest OpenAI model for my use case?▼

The Skill fetches the official latest-model guide from developers.openai.com first, which maps use cases like coding, realtime voice, image generation, and embeddings to current model IDs. If the remote guide is unavailable, it falls back to a bundled reference and discloses that.

How do I upgrade my app from gpt-4o to a newer OpenAI model?▼

The upgrade workflow inventories your model usage, classifies the source model family, and recommends either a model-string-only change or a light prompt rewrite. It preserves your existing reasoning effort settings and marks upgrades as blocked if they require API-surface or code changes.

What happens if the OpenAI docs MCP server is not installed?▼

The Skill attempts to install it with the codex mcp add command, retrying with escalated permissions if needed. Only if installation fails does it ask the user to run the command and restart, then falls back to web search restricted to official OpenAI domains.

Does the upgrade workflow change my API code or tool definitions?▼

No. The upgrade scope is limited to model-string replacements and directly related prompt edits. Any migration requiring API-surface changes, parameter rewrites, tool rewiring, or SDK changes is reported as blocked and out of scope.

When should I not rely on the bundled reference files?▼

Bundled references like latest-model.md and prompting-guide.md are fallbacks that can drift from current docs. The Skill always prefers live OpenAI documentation and treats official docs as the source of truth when conflicts arise.

Related Skills

Explore More Software Engineering Skills→
SoufianouassifSoufianouassif

ai-integrations-openai

Route OpenAI API calls through Replit AI Integrations without a personal API key.

Community
Advanced
tanveerriaztanveerriaz
1

llm-eval-golden-set

Builds a golden set of test cases and an eval runner to detect LLM output regressions.

Community
Intermediate
WrongStackWrongStack
224

prompt-engineering

Refine system prompts, tool descriptions, and skill definitions for WrongStack's 4-layer LLM prompt structure.

Official
Advanced
LannieYooLannieYoo

ai-llm_models

Provides LLM consultant assistance including selection, comparison, usage guidance, and troubleshooting.

Community
Intermediate