AI Prompt Engineering Lab

AI Prompt Engineering Lab

Rule: Do not store historical AI outputs or chat logs here. Only store reusable prompt templates.

This vault holds reusable prompt templates for the Testo project (testo-core). Each block below is a copy-paste starting point—a "Mega-Prompt" you can drop into Cursor or another AI assistant when starting a recurring task.

Before adding templates, read the Agent Context Guide for vault navigation rules. Prompts here should respect CLI contracts in Command Reference and execution patterns in QA Strategies.

How to use

1. Find the category that matches your task.
2. Copy the prompt inside the text code block (or replace the placeholder with your proven prompt).
3. Fill in any {placeholders} before sending.
4. When a prompt proves successful, paste it back into the matching block and give the subsection a descriptive title.

---

Documentation Generation

Template: Update docs after a code change

<!-- Paste your mega-prompt here -->

Template: Architecture note from source code

<!-- Paste your mega-prompt here -->

---

Code Refactoring

Template: Safe refactor with backward compatibility

<!-- Paste your mega-prompt here -->

Template: Reduce duplication (DRY / SOLID)

<!-- Paste your mega-prompt here -->

---

Testing & QA

Template: Add or extend unit tests

<!-- Paste your mega-prompt here -->

Template: E2E / cycle validation

<!-- Paste your mega-prompt here -->

---

CLI & Architecture

Template: New or changed CLI command

<!-- Paste your mega-prompt here -->

Template: Execution lifecycle / adapter change

<!-- Paste your mega-prompt here -->

---

CI/CD & Release

Template: Pipeline or runner integration

<!-- Paste your mega-prompt here -->

Template: Release checklist walkthrough

<!-- Paste your mega-prompt here -->

---

Debugging & Troubleshooting

Template: NDJSON / exit-code investigation

<!-- Paste your mega-prompt here -->

Template: Artifact or report pipeline debug

<!-- Paste your mega-prompt here -->

---

Related

• Agent Context Guide — master map for AI agents
• Command Reference — CLI contracts prompts should respect
• QA Strategies — how runs are triggered and logged
• CI-CD Pipeline Setup — CI wrapper patterns
• Technical Debt Tracker — code-level backlog (not prompt drafts)

---

Context & Links: UQO Engineering Hub, Index

---

Generated from docs/Archive/AI Prompt Engineering Lab.md. Edit it there; changes made in the wiki are overwritten.