candango-uat

Files (3)

• SKILL.md
• openai.yaml · agents/
• uat-template.md · references/

SKILL.md (first lines)

# Candango UAT

Create UATs that prove business behavior, not internal implementation.

All user-facing communication and written UAT reports must use `$caveman`.

## Inputs

Read:

- business definitions from conversation/context
- feature plan
- issue acceptance criteria
- design artifacts from `candango-design` when interface scope exists
- `docs/features/<feature-slug>/context.md`
- root `CONTEXT.md` as reference input
- product/domain docs

## Workflow

### 1. Extract Business Rules

List externally visible rules:

- actors
- triggers
- happy paths
- edge cases
- permission boundaries
- error states
- data visibility
- audit/notification outcomes
- visual/interface states when UI scope exists

### 2. Generate Scenarios

Use Given/When/Then format. Each UAT must map to at least one acceptance criterion or business rule.

Classify:

- critical
- important
- optional

Mark whether scenario is automatable, manual, or both.

### 3. Attach To Work

Save by default:

`docs/features/<feature-slug>/uat.md`

Keep all feature docs and HTML in `docs/features/<feature-slug>/`.

For each implementation issue, list related UAT IDs. If tracker supports subtasks, create UAT tasks only when user asks; otherwise keep UATs in docs and link from issues.

### 4. Validation Use

During autonomous execution, validator uses UATs as business oracle. Executor may add automated tests that cover UATs, but validator decides whether behavior matches.

## Output

Use `references/uat-template.md`.

From lucasbayma/skills, MIT.

More from lucasbayma/skills