Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations, help choosing the latest model for a use case, or model upgrade and prompt-upgrade guidance; prioritize OpenAI docs MCP tools, use bundled references only as helper context, and restrict any fallback browsing to official OpenAI domains.
90
87%
Does it follow best practices?
Impact
92%
1.05xAverage score across 3 eval scenarios
Critical
Do not install without reviewing
Narrow upgrade with explicit model target
Target preserved
100%
100%
Production code updated
100%
100%
Tests left unchanged
100%
100%
Eval baseline left unchanged
100%
100%
Docs left unchanged
28%
42%
No SDK migration
100%
100%
Upgrade class stated
75%
100%
Report: model usage inventory
71%
100%
Report: model-string updates
100%
100%
Report: reasoning-effort handling
0%
0%
Report: compatibility check
28%
100%
Report: validation section
85%
100%
Latest model resolution via script and upgrade classification
Script invocation logged
100%
90%
Guide URLs referenced
100%
100%
Fallback disclosure
100%
100%
Upgrade class stated
100%
100%
Model string updated
100%
100%
Narrow scope
100%
100%
Log: model usage inventory
100%
100%
Log: reasoning-effort section
100%
100%
Log: compatibility assessment
100%
100%
Log: validation plan
100%
100%
Documentation lookup source discipline and citation
Official sources only
70%
80%
No invented parameters
73%
86%
No unsourced capability claims
73%
86%
Paraphrase over verbatim
100%
90%
Fallback disclosed
100%
100%
Sources section present
100%
100%
Research log present
100%
100%
No speculation on unknowns
75%
91%
724cd51
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.