fix-bug
Arguments: $BUG_DESCRIPTION
You are a QA engineer and developer working together.
STEP 1 — UNDERSTAND THE BUG
Read $BUG_DESCRIPTION carefully. If insufficient detail, ask: "What were you doing when this happened? What did you expect to see? What did you see instead?"
STEP 2 — REPRODUCE FIRST (non-negotiable)
Before fixing anything: Write a Playwright test to tests/e2e/regression/[bug-id].spec.ts that replicates the exact user journey that triggers the bug. Run it. It must be RED (failing) before you touch the implementation.
If you can't make it fail, you don't understand the bug yet. Do not proceed until the test is RED.
More from ajaywadhara/agentic-sdlc-plugin
test-ui
Interactive browser testing via Playwright MCP. Navigates the live app, tests every screen, flow, and viewport — then generates permanent .spec.ts test files.
2start
Turn a vague idea into a structured PRD through guided conversation. The agent asks questions — you just talk.
2research
Market reality check — find competitors, read 1-star reviews, identify your angle. Run after /start.
2qa-run
8-agent QA loop: browser exploration via Playwright MCP, then analyze, plan, test, audit, heal, expand, snapshot. Quality gate score >= 85 to pass.
2build
TDD feature build loop: spec (RED) → implement (GREEN) → refactor. Pass the feature name as argument.
2coverage-review
Find untested areas across the codebase and fill coverage gaps. Run weekly or when the test suite feels stale.
2