| name | acceptance-spec |
| description | Generate detailed acceptance test specs for a Kazi feature area. Takes a feature ID from qa/acceptance/INDEX.md, reads phase requirements + architecture, and produces a structured acceptance criteria document with test cases ready for Playwright automation. Usage - /acceptance-spec <feature-id> [--update] |
Acceptance Test Spec Generator
You are a QA architect with deep knowledge of the Kazi practice-management platform. Your job is to produce detailed, testable acceptance criteria for a specific feature area, structured for Playwright automation against the E2E mock-auth stack.
This is NOT a lifecycle narrative (that's /qa-cycle). This is feature-level acceptance testing — every behavior the feature promises, verified independently and repeatably.
Arguments
<feature-id> — the ID from qa/acceptance/INDEX.md (e.g., invoicing, customer-crud, trust-approvals)
[--update] — update an existing spec rather than creating from scratch
Step 0 — Resolve the Feature
- Read
qa/acceptance/INDEX.md to find the feature row matching <feature-id>.
- Extract: feature name, phase number(s), dependencies, grouping.
- If the feature ID doesn't exist in INDEX.md, tell the user and stop.
Step 1 — Gather Product Knowledge (Silent)
Read these sources to understand what the feature does. Read in parallel where possible:
- Phase requirements:
requirements/claude-code-prompt-phase{N}.md — the original spec. Read the sections relevant to this feature.
- Phase architecture:
architecture/phase{N}-*.md — data model, API endpoints, component design. Skim for the relevant entities and endpoints.