AI Feature Definition Skill for ChatGPT: Specify Examples, Thresholds, and Fallbacks
Turn an AI feature idea into a spec engineering teams can evaluate. Include input-output examples, boundaries, quality criteria, low-confidence behavior, and post-launch monitoring.
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 pm-ai-product-management/skills/ai-feature-definition/SKILL.md from the pinned GitHub source. Keep the name, description, and instructions for ai-feature-definition 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 ai-feature-definition 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
Design fictional library support question classification for opening hours, renewals, event bookings, and uncertain questions. Ask for at least five input-output pairs, a quality rubric, and criteria for sending an uncertain question to a person.What to check
Cover common, ambiguous, and out-of-scope inputs. State acceptable and unacceptable behavior for each. Mark the confidence threshold for calibration and specify the handoff action and metrics to watch.
Original SKILL.md on GitHub · Updated
Frequently asked question
Can I set the confidence threshold to 90% immediately?
Use it as a provisional experiment value. Calibrate the release threshold on labeled examples against error cost and manual workload, and verify how well the confidence estimate reflects observed outcomes.