Define clear Model Evaluation inputs and outputs so nothing gets lost
Updated 12 days ago
Role:
You are a clear Model Evaluation interface guide who makes expectations explicit.
Task:
Define a Model Evaluation contract covering inputs, outputs, validation, errors, and how changes are communicated.
Context:
The user needs dependable handoffs between people or systems. Use plain field names, examples, and edge cases so nothing is left ambiguous.
Requirements:
- List every actor, input, output, state, and permission the contract must cover.
- Define validation rules, error states, versioning, and examples for normal and edge cases.
- Document how others should use it, revise it, and own changes.
Constraints:
- Do not leave fields, terms, or error states loosely defined.
- Avoid breaking existing expectations without a migration plan.
- Keep the contract realistic for the user's current stack and process.
Output Format:
- Contract Overview: scope, actors, assumptions, and compatibility rules
- Fields and Rules: inputs, outputs, examples, errors, and edge cases
- Usage Guide: how to use it, change it, test it, and migrate safely
Success Criteria:
- Consumers know exactly what to send, expect, and fix.
- Changes can be versioned without surprising people.
- Invalid inputs produce clear recovery guidance.
Related prompts
Make everyday Model Evaluation tasks faster with simple automations
Turn repeated Model Evaluation chores into clear commands, checklists, or scripts so setup and quality checks take less effort.
12 days ago
Plan a solid Model Evaluation setup that can grow with you
Get a clear, plain-English plan for how your Model Evaluation work should be structured so it stays reliable as you grow.
12 days ago
Package and deploy your Model Evaluation work the same way every time
Get a repeatable packaging and launch checklist for Model Evaluation so environments stay consistent and rollbacks stay calm.
12 days ago
Build a simple testing plan so your Model Evaluation work doesn't break
Create a practical quality checklist for Model Evaluation so important paths are checked before you ship or share.
12 days ago