{"slug": "quality-bar-md", "title": "quality-bar.md", "summary": "A developer has published a template called 'quality-bar.md' for teams to define quality criteria before building with AI. The template includes sections for purpose, user context, behavioral specification, and ethical boundaries, emphasizing that the conversation it forces is the work itself.", "body_md": "| # quality-bar.md | |\n| A template for encoding your quality criteria before building with AI. | |\n| Fill this in as a team before the first prompt is written. | |\n| The conversation it forces is the work. | |\n| Update it as your understanding evolves. Version it. | |\n| This is a living document, not a one-time exercise. | |\n| --- | |\n| ## 1. Purpose | |\n| **What problem does this solve?** | |\n| One sentence. The real problem, not the feature description. | |\n| **Who experiences this problem, and when?** | |\n| Be specific. Not \"users\" — a real person in a real moment. | |\n| **What would they say if this worked perfectly?** | |\n| Write the sentence they would say to a colleague the next day. | |\n| **What would they say if it failed?** | |\n| Write the complaint. Be honest about what failure looks like from their side. | |\n| --- | |\n| ## 2. The user in the moment | |\n| **What state is the user in when they encounter this?** | |\n| Rushed, confused, expert, anxious, curious? The emotional and cognitive state matters. | |\n| **What do they know that the system doesn't?** | |\n| What context, history, or preference exists in their head that AI cannot see? | |\n| **What do they assume the system can do that it can't?** | |\n| Where is the gap between user expectation and actual system capability? | |\n| --- | |\n| ## 3. Behavioral specification | |\n| **In normal conditions, the AI should:** | |\n| **When it doesn't know, the AI should:** | |\n| **When it is uncertain, the AI should:** | |\n| **When the user is wrong, the AI should:** | |\n| **When it makes a mistake, the AI should:** | |\n| --- | |\n| ## 4. What good looks like | |\n| **Describe a perfect interaction, step by step.** | |\n| **What does the output look, sound, and feel like when it's right?** | |\n| **What would a senior designer say about it in critique?** | |\n| What specific quality markers would they call out? | |\n| --- | |\n| ## 5. What wrong looks like | |\n| **Describe an output that looks right but is subtly wrong.** | |\n| The failure mode that passes a visual check but fails the intent check. | |\n| **Describe an output that is confidently wrong.** | |\n| The hallucination or overreach case specific to your product. | |\n| **Describe an output that is technically correct but unhelpful.** | |\n| The case where the AI answered the question but missed the point entirely. | |\n| --- | |\n| ## 6. Voice and tone | |\n| **How does this AI speak?** | |\n| Three adjectives that describe the voice. Three it explicitly avoids. | |\n| Sounds like: | |\n| Never sounds like: | |\n| **What does on-brand output look like versus off-brand?** | |\n| A specific example of each for this product. | |\n| **What does the AI never say, even if prompted?** | |\n| --- | |\n| ## 7. The not-doing list | |\n| **This product refuses to:** | |\n| **This product will never do, even if asked:** | |\n| **We have considered and deliberately declined:** | |\n| List the things that would technically work but that you are choosing not to do, and why. | |\n| --- | |\n| ## 8. Ethical boundaries | |\n| **What user autonomy must this always preserve?** | |\n| **What data does this handle, and what does it never do with it?** | |\n| **Where might this cause harm, and how have we addressed it?** | |\n| **Who could be disadvantaged by this, and what have we done about it?** | |\n| --- | |\n| ## 9. Criteria for AI-generated output | |\n| **Every output must:** | |\n| **An output fails the bar if it:** | |\n| **Who reviews output, and what are they specifically looking for?** | |\n| **At what stage does review happen? Who has authority to reject and send back?** | |\n| --- | |\n| ## 10. How we know | |\n| **What user signal tells us the bar is being met?** | |\n| **What signal tells us quality has slipped?** | |\n| **How often do we review these criteria, and who owns that review?** | |\n| --- | |\n| ## Metadata | |\n| Last updated: | |\n| Updated by: | |\n| Version: | |\n| Approved by: | |\n| --- | |\n| Embed this file in your tooling so it travels with the work. | |\n| This document should live where the work lives, not in a folder nobody opens. | |\n| MIT Licensed — Owl-Listener (MC Dean) |", "url": "https://wpnews.pro/news/quality-bar-md", "canonical_source": "https://gist.github.com/Owl-Listener/38b62567a5bb2806b72fec9a6f0d76b6", "published_at": "2026-06-14 17:05:55+00:00", "updated_at": "2026-07-11 07:38:38.198862+00:00", "lang": "en", "topics": ["ai-ethics", "developer-tools", "ai-safety"], "entities": [], "alternates": {"html": "https://wpnews.pro/news/quality-bar-md", "markdown": "https://wpnews.pro/news/quality-bar-md.md", "text": "https://wpnews.pro/news/quality-bar-md.txt", "jsonld": "https://wpnews.pro/news/quality-bar-md.jsonld"}}