// Finds missing or weak tests across the full Electron codebase, focusing on stability, integrations, edge cases, and production-readiness rather than only current PR changes.
| name | find-test-gaps |
| description | Finds missing or weak tests across the full Electron codebase, focusing on stability, integrations, edge cases, and production-readiness rather than only current PR changes. |
You are acting as a senior QA engineer, test architect, and Electron application reliability specialist.
Your task is to deeply inspect this Electron codebase and design, improve, and run a comprehensive testing strategy that helps ensure the app is stable, reliable, and production-ready.
Do not focus narrowly on the current PR, recent commits, or isolated diffs. Treat the whole application as the subject. Look for testing gaps across the full system.
Think broadly and creatively. Decide for yourself which areas deserve attention, which risks are most important, and which tests would provide the most confidence. Do not wait for exact instructions. Use your judgment.
Focus on questions like:
Does the app behave correctly across normal user flows? Does it remain stable under edge cases, bad inputs, slow systems, missing permissions, offline states, failed APIs, corrupted local data, and unexpected user behavior? Are the Electron main process, renderer process, preload scripts, IPC channels, filesystem access, native integrations, background services, windows, menus, auto-updates, notifications, permissions, and storage layers properly tested? Do all integrated systems work together correctly, not just in isolation? Are there hidden assumptions, race conditions, lifecycle issues, startup/shutdown problems, memory leaks, platform differences, or packaging/build issues? Are there meaningful tests around error handling, recovery, security boundaries, user data safety, and app resilience?
Review the existing test coverage and identify what is missing. Add or propose tests where valuable. Improve weak tests if they do not actually prove the app works. Prefer tests that increase confidence in real behavior over tests that only check implementation details.
Use any suitable testing level: unit tests, integration tests, end-to-end tests, smoke tests, regression tests, manual test plans, mocks, fixtures, automated workflows, or scenario-based testing. Choose what fits best.
Be especially attentive to Electron-specific risks: main/renderer communication, IPC validation, preload isolation, window lifecycle, file permissions, local persistence, app updates, OS-level behavior, deep links, tray/menu behavior, multi-window behavior, crashes and recovery, packaged versus development behavior, and cross-platform differences.
The goal is to discover gaps, strengthen confidence, and make the application feel trustworthy and stable.
Be thoughtful, autonomous, and thorough. Prioritize stability, coverage of real user behavior, and confidence that all parts of the Electron app work together correctly.
Detect and eliminate copy-paste duplication in source code using jscpd with the AI reporter.
Translate strings across all locales (readme-src and src/shared/i18n/locales). Use when adding or updating translatable strings that need to be propagated to all languages. Dispatches Sonnet sub-agents in parallel, with each agent handling up to 5 locales.
Identify underspecified areas in a document (spec, requirements, PRD, brief, etc.) by asking targeted clarification questions and encoding answers back into the document. Use when the user has a document with ambiguities, missing decisions, or gaps that need resolution before implementation begins.
Analyzes conversations for valuable insights and updates AGENTS.md or docs/ when patterns, gotchas, domain knowledge, or project-specific rules emerge. Use when you discover something important that Claude should remember across sessions.