القائمة
Use for improving LawnBerry Pi AI result quality without breaking the existing backend contract. Covers ai_service review, model artifact or rule improvement, conservative evaluation, performance notes, and avoiding premature accelerator-specific churn.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Use for LawnBerry Pi simulation vs real-hardware validation. Covers SIM_MODE selection, .env and config checks, simulation-safe test-first workflow, hardware preflight expectations, and avoiding false claims that local success proves on-device behavior.
Review regression-sensitive LawnBerry Pi manual control and camera paths: RoboHAT USB handoff, watchdog feeding, motor authorization, joystick responsiveness, MJPEG and snapshot fallback, camera ownership, and stream backpressure handling.
Keep LawnBerry Pi maintainer documentation synchronized with implementation. Use when runtime behavior, supported hardware scope, subsystem maturity, immediate focus, two-week plan, release notes, or callable interfaces have changed and docs must be updated in the same pass.
Use when returning to LawnBerry Pi after time away, starting substantial maintenance, or needing a fast project re-entry workflow. Covers docs to read first, runtime ports, hardware baseline, simulation vs hardware mode, important files by task, and validation planning.
Use when a LawnBerry Pi task spans re-entry, specialist routing, implementation, documentation sync, and validation. Covers choosing the right specialist workflow and sequencing investigation, changes, doc updates, and test planning.
Use for LawnBerry Pi mission persistence and recovery work. Covers what mission state should survive restart, safe restart semantics, paused vs running recovery behavior, lifecycle validation, and persistence-backed tests.
| name | ai-model-quality-pass |
| description | Use for improving LawnBerry Pi AI result quality without breaking the existing backend contract. Covers ai_service review, model artifact or rule improvement, conservative evaluation, performance notes, and avoiding premature accelerator-specific churn. |
| argument-hint | What AI quality improvement should be made behind the current backend contract? |
| user-invocable | true |
This skill focuses AI work on better quality behind the existing API rather than reopening the contract every time perception needs improvement.
backend/src/services/ai_service.pytests/test_ai_api.pytests/unit/test_ai_service.py