home/writing/prompt-quality-standard

PROMPT QUALITY STANDARD

GPTClaudeGemini··▲ 1,112 copies·updated 2026-07-14

prompt-quality-standard.prompt

# Prompt Quality Standard — Sentra Prompt 2026

**Scope:** `apps/internal/sentra-prompt`  
**Authority:** `lib/prompt-quality/contract.ts` (runtime SSOT)  
**Last updated:** 2026-06-04

---

## 1. What is a "Good Prompt"?

A good prompt is **self-contained, unambiguous, and testable**. Anyone — human or LLM — should be able to read it cold and produce the intended output without asking follow-up questions.

Four questions every prompt must answer:

| Question | Field |
|---|---|
| Who is the agent? | `role` |
| What must be done? | `task` |
| What context matters? | `context` |
| What does good output look like? | `constraints` + `formatSpec` |

If any answer is missing, the prompt is incomplete.

---

## 2. The Super Prompt Schema (SSOT)

All structured prompts in Sentra Prompt conform to `SuperPromptSchema` in `lib/prompt-quality/contract.ts`:

when to use it

Community prompt sourced from the open-source GitHub repo drferdii/Myprompt (no explicit license). A "PROMPT QUALITY STANDARD" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.