Writing Skills for ChatGPT: Test Workflow Instructions With Baselines and Counterexamples
Start from an observed failure and write a testable workflow. Define behavior with and without the skill and include a case where it should stay inactive.
中文Install in ChatGPT
Use an eligible ChatGPT workspace with permission to create skills and, for uploads, upload permission. The installation path for this text-only adaptation was checked against OpenAI documentation. Workspace runtime verification is pending.
- Download .claude/skills/writing-skills/SKILL.md from the pinned GitHub source. Keep the name, description, and instructions for writing-skills together as a standalone text skill.
- In a ChatGPT workspace with skills and uploads enabled, open Plugins, Skills, Create, and Upload from your computer. Choose a skill package format accepted by the file picker. You can also choose Create with editor and enter the original name, description, and instructions.
- After the scan completes, check that writing-skills appears in Installed. Type @ in a new chat, select the skill, and use the fictional example on this page to check the output.
Example task
A fictional course directory assistant drops course duration when structuring lists. Use the same three records to plan a baseline, expected behavior with the skill, and an out-of-scope task.What to check
The baseline records any dropped durations. The positive case checks all three durations and unknown fields, while an unrelated task checks false activation. Outcomes are left for the run.
Original SKILL.md on GitHub · Updated
Frequently asked question
Does one successful run establish stability?
Repeat with representative and boundary cases. Keep inputs comparable and record expected and actual behavior so later revisions can be checked.