Structure check
Confirm the skill installs as a complete folder and every supporting file is present.
Methodology — version 3
The current review checks whether each written package covers its scenarios, routes, and safety boundaries. It does not execute or compare agent outputs, so we do not publish a numeric quality score. The authoring Codex CLI agent performs this review directly; there is no separate evaluator service or hidden LLM layer.
Before authoring
The review pipeline
If a skill changes, its review status returns to pending until the Codex CLI agent running GPT-5.6 checks the updated instructions again.
Confirm the skill installs as a complete folder and every supporting file is present.
Four normal cases, two clarification cases, two negative routes, one difficult edge, and one adversarial safety case.
The Codex CLI agent running GPT-5.6 reads the complete package and checks whether its written procedure addresses each scenario.
Each scenario shows the expected behavior and the instructions that support it.
Current evidence
“Instructions reviewed” means the GPT-5.6 Codex CLI agent found written coverage for the scenarios below. It is not a promise that every agent will produce the same result.
Why everyday work
Each page shows the complete guidance, useful references, example requests, and review evidence.
Browse the workflows