Token导航 LogoToken导航

Agent Skill · 效率

medication-channel(用药频道)

medication-channel 用于补充效率相关能力,适合在 OpenClaw 中需要让 Agent 承接效率相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

medication-channel 是效率类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描已通过,按安装量在效率类的 21,003 个 Skill 里排第 17,210。

总安装 2,661周安装 112GitHub Stars 公开资料未说明下载 932核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:medication-channel(用药频道)
来源仓库:https://github.com/cdmichaelb/medication-channel
安装命令:
openclaw skills install medication-channel
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install medication-channel

这个 Skill 是做什么的

效率类 Skill 把重复的整理、汇总、转换工作交给助手完成,减少手工复制粘贴。

medication-channel 通过确定性路由机制处理药物事件,提升用药跟踪的准确性。适合需要自动化管理处方流转、药品提醒和用药记录的临床或家庭场景。输入药物名称、剂量或使用时间,系统按预设规则分配处理通道并返回执行状态。使用前需确认权限范围和集成方式,建议查阅原始仓库了解配置细节。

该技能不涉及药物推荐,仅辅助流程管理,重要用药请咨询专业医师。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install medication-channel;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

medication-channel 怎么安装?

在终端执行 openclaw skills install medication-channel。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

medication-channel 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

medication-channel 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

medication-channel 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

medication-channel 用的人多吗?

公开数据是累计安装 2,661 次,近一周安装 112 次,下载 932 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

medication-channel 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理效率类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

效率类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
github

GitHub 代码协作 · OpenClaw

7,632,918545需要联网MIT-0通过
nano-pdf

Nano PDF 文档 · OpenClaw

4,249,476231需要联网MIT-0通过
auto-updater

自动更新程序 · OpenClaw

2,964,063386操作浏览器MIT-0通过
video-frames

视频帧 · OpenClaw

2,621,304112需要联网MIT-0通过
gemini

Gemini 助手 · OpenClaw

2,547,39648需要联网MIT-0有提醒
medication-channel

用药频道 · 本页

2,661—接触敏感数据MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
medication-channel
version
1.0.0
description
Handle messages in a medication-tracking channel by routing medication events through a deterministic medication tracker instead of freehand model judgment. Use when working in a dedicated medication channel, especially for logging taken/missed/extra medication events, morning/evening completion, timestamp-grounded confirmations, and avoiding hallucinated reminders or fake times.
env
MEDICATION_TIMEZONE
description
IANA timezone for local display (e.g. America/Los_Angeles)
required
true
WORKSPACE
description
Root workspace directory for data files
required
false

Medication Channel

Use the deterministic medication logging path whenever source message metadata is available.

Configuration

Set the following environment variables or replace placeholders before use:

  • MEDICATION_TIMEZONE — IANA timezone string for local time display (e.g. America/Los_Angeles, Europe/Berlin)
  • WORKSPACE — root directory for data files (defaults to ~/.openclaw/workspace)

Medication names and schedules are configured at the top of scripts/tracker_v2.py — edit the MORNING_MEDS, EVENING_MEDS, and KNOWN_MEDS lists to match your regimen.

Core Rules

  • Treat the configured channel as the dedicated medication channel.
  • Prefer the log_from_discord.sh wrapper or tracker_v2.py log-from-message.
  • Never use now as a logged/displayed time.
  • Never derive event time from the model turn time.
  • Use the source Discord message timestamp unless the user explicitly gives a different time.
  • Convert displayed times to the configured timezone.
  • Never log reminder chatter as medication history.
  • Never treat an assistant reminder message, assistant follow-up, assistant summary, or assistant self-referential chatter as a medication event.
  • Never emit a confirmation like "taken", "logged", or "recorded" in response to an assistant-generated reminder unless a real user medication report was just logged through the script path.
  • Never claim something was logged unless the script path actually ran.
  • Prefer raw-message passthrough to the tracker instead of model-side parsing.

Preferred Workflow

Log a medication message from Discord

Use when the user reports:

  • taken / took
  • done / completed
  • missed / skipped
  • extra dose
  • clear natural-language medication intake

Preferred path:

scripts/log_from_discord.sh \
  "$MESSAGE_TEXT" \
  "$CHANNEL_ID" \
  "$MESSAGE_ID" \
  "$AUTHOR_ID" \
  "$TIMESTAMP_UTC"

Equivalent direct call:

python3 scripts/tracker_v2.py log-from-message \
  "$MESSAGE_TEXT" \
  "$CHANNEL_ID" \
  "$MESSAGE_ID" \
  "$AUTHOR_ID" \
  "$TIMESTAMP_UTC"

Then use the returned row or format-confirmation output for the confirmation.

Parse only

Use only when inspecting behavior or debugging.

python3 scripts/tracker_v2.py parse "$MESSAGE_TEXT" "$TIMESTAMP_UTC"

Format a confirmation

Use when you already have the JSON row and want the standard confirmation string.

python3 scripts/tracker_v2.py format-confirmation '<json-row>'

Special Guardrail: Reminder Messages

If the message being handled is itself a reminder, or the user is replying to/quoting a reminder message, do not assume medication was taken.

Examples that are not medication events by themselves:

  • an assistant message like Reminder: it's time to take your evening medication.
  • an assistant message like Evening medication taken (recorded at ...)
  • a user message like something went wrong again
  • a user reply that is discussing a reminder failure rather than reporting intake

In these cases:

  • do not call the tracker unless the user message itself contains an actual medication event
  • do not fabricate a confirmation
  • debug/explain briefly instead

What the Tracker Already Handles

The tracker is preferred because it already handles:

  • common natural-language medication phrases
  • normalization of common misspellings / aliases
  • morning/evening inference
  • dedupe by source Discord message id
  • local timezone display formatting
  • structured logging

So do not manually re-implement those steps unless debugging the tracker itself.

If Uncertain

Ask a short clarifying question only when needed, such as:

  • whether the message is a real medication event vs general discussion
  • whether the user is saying they skipped something or are uncertain
  • whether a custom explicit time should override the source timestamp

Data Access

This skill reads and writes persistent files in the workspace:

  • Writes $WORKSPACE/data/medication_log_v2.csv — one row per medication event (med name, status, timestamps, source message ID)
  • Writes deduplication state keyed by source message ID
  • Reads medication config from MORNING_MEDS/EVENING_MEDS/KNOWN_MEDS lists at top of scripts/tracker_v2.py
  • No network calls — fully local, stdlib-only Python
  • No external APIs — no vision, no remote services

Required Files

The following files are included in the bundle:

  • scripts/log_from_discord.sh — Discord wrapper script
  • scripts/tracker_v2.py — medication parser/logger
  • references/CHANNEL_RULES.md — optional channel behavior reference
  • references/README.md — optional setup reference

平台分布

OpenClaw

73.85%
按下载量换算688

安全审计

VirusTotal

未展示

ClawScan

通过

Static analysis

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills