Agent Skills › Fullive-AI/Anima › humidifier

humidifier

GitHub

用于智能家居环境中加湿器/除湿器的设备级湿度控制与偏好学习，涵盖舒适目标、季节启发式规则及开关模式等操作决策。

skills/system/humidifier/SKILL.md Fullive-AI/Anima

触发场景

用户询问或设置室内湿度目标需要根据季节或睡眠场景调整湿度策略执行加湿器或除湿器的启停、模式切换操作

安装

npx skills add Fullive-AI/Anima --skill humidifier -g -y

更多选项

非标准路径

npx skills add https://github.com/Fullive-AI/Anima/tree/main/skills/system/humidifier -g -y

不安装直接使用

npx skills use Fullive-AI/Anima@humidifier

指定 Agent (Claude Code)

npx skills add Fullive-AI/Anima --skill humidifier -a claude-code -g -y

安装 repo 全部 skill

npx skills add Fullive-AI/Anima --all -g -y

预览 repo 内 skill

npx skills add Fullive-AI/Anima --list

SKILL.md

Frontmatter

{
    "name": "humidifier",
    "metadata": {
        "version": "0.1.0",
        "device_types": [
            "humidifier",
            "dehumidifier"
        ]
    },
    "description": "Use when reasoning about humidifiers or dehumidifiers in Anima. Covers comfort humidity targets, seasonal heuristics, and choosing `set_humidity`, `set_mode`, `turn_on`, or `turn_off`."
}

Humidifier

Use this skill for device-level humidity control and preference learning.

Load These Resources

references/knowledge.md for comfort bands, device interactions, and guardrails.
references/decide.md when generating a single-device action.
references/learn.md when summarizing long-term preference patterns.
scripts/actions.py for the structured action helpers exposed to the runtime.

Working Rules

Prefer gradual target changes over abrupt swings.
Treat water level, season, sleep context, and AC/heating interaction as first-order signals.
Return structured actions only from the supported action set in scripts/actions.py.