菜单
Run Wildlife Watcher ruff, mypy, pytest, coverage, and ML pipeline checks through the repo harness. Use for lint, typecheck, regression, quality gate, and pre-merge validation tasks.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Verify that every required Wildlife Watcher directory has an AGENTS.md file. Use for agent customization maintenance, repo hygiene, and hierarchy validation.
Build the Wildlife Watcher camera-node firmware with PlatformIO and inspect the resulting compile output. Use for firmware compile checks, dependency issues, and binary-size review.
Publish synthetic Wildlife Watcher detection payloads to the configured MQTT broker. Use for kiosk smoke tests, UI validation, MQTT topic checks, and display-node debugging without hardware.
Bootstrap a dedicated Wildlife Watcher roadmap worktree for Phase 2 PIR/power or Phase 3 ML work. Use for worktree setup, branch isolation, and phase-scoped experimentation.
| name | run-quality-gates |
| description | Run Wildlife Watcher ruff, mypy, pytest, coverage, and ML pipeline checks through the repo harness. Use for lint, typecheck, regression, quality gate, and pre-merge validation tasks. |
| argument-hint | Optional dry-run flag or reason for running the quality gate. |
| user-invocable | true |
python .agents/harness/orchestrator.py quality.ml-pipeline/, also run python .agents/harness/orchestrator.py ml-pipeline-lint, ml-pipeline-typecheck, ml-pipeline-test, and ml-pipeline-smoke.