What problem does it solve? After an automated implementation stage, frontend components often remain as stubs — placeholder text, unconnected hooks, dead links, and orphan components — while appearing complete. This Skill systematically detects and fixes those gaps so every feature in the PRD is genuinely functional before QA begins. ## Core Features & Use Cases - Automated Stub Scanning: Runs grep-based scans for placeholder text, thin components, orphan files, dead onClick handlers, mock data, and missing loading/error states across the codebase. - Size-Based Issue Classification: Categorizes findings into Category A (quick fixes), Category B (medium fixes), and Category C (escalations requiring re-implementation), with a hard gate decision (PASS, CONDITIONAL PASS, FAIL-RETRY, FAIL-ESCALATE). - Multi-Agent Fix Coordination: Spawns frontend, backend, and end-to-end verification teammates with mutually exclusive file assignments to fix issues and validate results. - Use Case: After an AI pipeline implements all PRD features, run this audit to find that a dashboard widget shows hardcoded zeros and a settings form has no submit handler, then automatically fix both and produce a gate report. ## Quick Start Ask the agent to run the implementation completeness audit for Stage 5B against the PRD in plancasting/prd and report the gate decision.
