← 返回 Skills 市场
Ping Me
作者
Yuyang Shen
· GitHub ↗
· v1.0.3
· MIT-0
311
总下载
1
收藏
0
当前安装
4
版本数
在 OpenClaw 中安装
/install ping-me
功能描述
One-shot reminders via natural language. Auto-detects channel and timezone. Say 'remind me...' in any language and get pinged when it's time. Works with ever...
安全使用建议
This package appears to do exactly what it says: create one‑shot reminders via openclaw cron. Before installing, ensure you trust the OpenClaw runtime that will execute the scripts, and confirm the gateway environment variables (OPENCLAW_SESSION_KEY / OPENCLAW_TO / OPENCLAW_CHANNEL) that the scripts read are safe to expose in your agent environment — those may contain session identifiers. The skill stores its own config.json (contains default channel, 'to' target, tz, emoji) under its baseDir; if you want to restrict persisted data, inspect or relocate that file. If you have concerns about the gateway injecting sensitive tokens into env vars, review how your gateway populates OPENCLAW_SESSION_KEY and OPENCLAW_TO before enabling the skill.
功能分析
Type: OpenClaw Skill
Name: ping-me
Version: 1.0.3
The 'ping-me' skill is a legitimate reminder utility that uses the OpenClaw platform's native cron functionality. The shell scripts (ping-me.sh, ping-me-config.sh, etc.) follow secure coding practices by using Python to handle JSON data via stdin and sys.argv, effectively mitigating shell injection risks. The skill only accesses environment variables provided by the OpenClaw gateway (e.g., $OPENCLAW_CHANNEL) and does not exhibit any signs of data exfiltration, unauthorized network access, or malicious prompt injection.
能力评估
Purpose & Capability
Name/description (reminders) match the included scripts and SKILL.md. The scripts only interact with OpenClaw (cron) and local config.json, which is proportionate to creating/delivering reminders. Required binaries (openclaw, python3) are appropriate.
Instruction Scope
Runtime instructions are narrowly scoped: parse time/message, call local scripts, and confirm. The SKILL.md and scripts reference gateway-provided environment variables (OPENCLAW_CHANNEL, OPENCLAW_TO, OPENCLAW_SESSION_KEY, OPENCLAW_SESSION_CHANNEL) to auto-detect channel/target; this is expected for a multi-channel reminder skill but is worth noting because the skill reads session env vars that may contain identifiers or tokens.
Install Mechanism
No install spec (instruction-only with included scripts). No downloads or archives; all code is contained in the package. Low install risk.
Credentials
The skill does not declare required env vars and requests no API keys. The scripts read gateway env vars (OPENCLAW_TO, OPENCLAW_SESSION_KEY, OPENCLAW_CHANNEL, OPENCLAW_SESSION_CHANNEL, TZ) to determine delivery target and channel. This is coherent with multi-channel delivery, but those env vars can contain session identifiers; users should be comfortable exposing those values to the skill's runtime environment.
Persistence & Privilege
always is false and the skill only writes to its own config.json under baseDir. It does not modify other skills or system-wide agent settings. It uses OpenClaw cron jobs for persistence, which is consistent with its purpose.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install ping-me - 安装完成后，直接呼叫该 Skill 的名称或使用
/ping-me触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.3
**1.0.3 Summary:** Adds delivery target (`--to`) support for multi-channel compatibility, along with new first-time setup guidance.
- Added `config.json` to persist user/channel/delivery settings.
- SKILL.md updated with instructions for `--to` option to set delivery target (e.g., `qqbot:c2c:<openid>`), resolving multi-channel announce requirements.
- Delivery target auto-detection detailed: via environment variables, session key, or config fallback.
- First-time setup troubleshooting steps included for users needing to configure channel or delivery target.
- No breaking changes to previous command flows; basic usage unchanged for existing channels.
v1.0.2
- Clarified response flow: Call the script first, then confirm with one concise sentence, no preamble or time calculation explanation.
- Removed strict warnings about forbidden/required agent behaviors; replaced with a simpler, user-focused style guide.
- Updated agent instructions and response style section for added clarity and brevity.
- No changes to core functionality or commands; documentation only.
v1.0.1
**ping-me 1.0.1 Changelog**
- Added README.md with detailed usage and instructions.
- Introduced `scripts/ping-me-config.sh` for viewing and changing reminder settings (timezone, channel, emoji, language).
- Improved SKILL.md: clarified that channel and timezone are now auto-detected; updated agent instructions for new config script.
- Users can now interactively view, update, or reset skill settings using the new script.
v1.0.0
ping-me 1.0.0
- Initial release: set one-shot reminders using natural language.
- Supports relative (e.g., "in 3 minutes") and absolute (ISO 8601) time formats.
- Works across multiple chat channels (QQ, Telegram, Discord, WhatsApp, etc.).
- Strict response rules: only one confirmation message after the reminder is set, no pre-reply or reasoning.
- Includes commands to list and cancel reminders.
元数据
常见问题
Ping Me 是什么？
One-shot reminders via natural language. Auto-detects channel and timezone. Say 'remind me...' in any language and get pinged when it's time. Works with ever... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 311 次。
如何安装 Ping Me？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install ping-me」即可一键安装，无需额外配置。
Ping Me 是免费的吗？
是的，Ping Me 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Ping Me 支持哪些平台？
Ping Me 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Ping Me？
由 Yuyang Shen（@ethan-shen-individual-lab）开发并维护，当前版本 v1.0.3。
推荐 Skills