← 返回 Skills 市场
Glancely
作者
JunjieBusiness
· GitHub ↗
· v0.2.0
· MIT-0
26
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install glancely
功能描述
Personal tracker framework. Scaffold new habit/workout/mood/reminder trackers from natural language. Dashboard, cron, all in one skill.
使用说明 (SKILL.md)
What glancely is
A personal tracker framework that puts all your habits in one dashboard. You describe what you want to track, and glancely scaffolds it — migrations, cron, dashboard panel — in one command.
Routing user intent
When a user asks for something, follow this dispatch table:
| User says | Action |
|---|---|
| "log mood" / "how I feel" | Check ~/.glancely/components/mood/. If exists → read SKILL.md → run log.py. If not → read examples/mood/SKILL.md for reference → scaffold first, then log. |
| "remind me" / "add reminder" | Check ~/.glancely/components/reminder/. Same fallback to examples/reminder/SKILL.md. |
| "what's my MIT" / "most important task" | Check ~/.glancely/components/mit/. Fallback to examples/mit/SKILL.md. |
| "log diary" / "track time" | Check ~/.glancely/components/diary_logger/. Fallback to examples/diary_logger/SKILL.md. |
| "build dashboard" / "show dashboard" | Run glancely dashboard build and open ~/.glancely/dashboard/index.html. |
| "create tracker" / "track my" / "scaffold a habit" | Read glancely/skills/scaffold_component/SKILL.md. Infer fields, cron, notifications from the user's description. Propose a plan. Confirm. Scaffold. |
| "what trackers do I have" | Run glancely list. |
Per-route workflow
- Check if the user already has a matching component in
~/.glancely/components/ - If yes: read its SKILL.md for field docs, then call its scripts/log.py
- If no: read the matching example in
examples/for reference, then use the scaffold flow to create it first - Report the result
Setup
If glancely CLI is missing or ~/.glancely/ does not exist:
- Run
pip install glancely - Run
glancely setup(minimal init — migrations only) - If user wants cron, ask for agent_id / session info and write
~/.glancely/openclaw.toml
Scaffold flow (via scaffold_component/SKILL.md)
- Analyze the user's goal. Infer tracker name, fields, cron schedule, notification text.
- Propose a plan — list trackers, fields, and cron before touching anything. Ask for confirmation.
- Scaffold each confirmed tracker:
glancely scaffold --name ... --field ... --cron ... --notify ... - Dashboard auto-builds after each scaffold.
安全使用建议
Before installing, verify the glancely package source, review any generated files under ~/.glancely, and be cautious about enabling cron or providing agent/session information. Use the skill only with tracker components you trust, and remove scheduled jobs or stored session config when you no longer need them.
功能分析
Type: OpenClaw Skill
Name: glancely
Version: 0.2.0
The skill acts as a framework that installs an external Python package ('glancely'), configures cron jobs for persistence, and requests sensitive OpenClaw 'agent_id' and session info. It employs a plugin-like architecture in SKILL.md that directs the agent to execute scripts (log.py) and follow instructions from locally stored markdown files (~/.glancely/components/), which introduces a risk of indirect prompt injection or arbitrary code execution if those local files are tampered with. While these features are aligned with the stated goal of a tracker framework, the combination of package installation, automated scheduling, and dynamic execution of local code represents a high-risk capability set.
能力评估
Purpose & Capability
The tracker, dashboard, scaffold, and cron features match the stated purpose, but they involve local file writes, command execution, and scheduled behavior.
Instruction Scope
The skill directs the agent to read local component SKILL.md files and then run component scripts from ~/.glancely, which can make persistent local/generated content influence future agent behavior and code execution.
Install Mechanism
The skill instructs installation of the glancely Python package via pip, which is purpose-aligned but unpinned and not reviewable in the provided artifacts.
Credentials
Running component scripts from a user-writable data directory is more powerful than simple tracking data storage and is not clearly bounded by validation, path restrictions, or per-run user approval.
Persistence & Privilege
Optional cron setup asks for agent_id/session info and writes a persistent OpenClaw config file, but the artifacts do not define what authority that session grants or how it is protected/revoked.
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 glancely - 安装完成后，直接呼叫该 Skill 的名称或使用
/glancely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
v0.2.0
- Added dispatch table to map natural language user requests to corresponding tracker actions and scaffolding flows.
- Introduced per-route workflow for consistent handling of existing and new tracker requests.
- Updated setup instructions for first-time users to install CLI and initialize data directory.
- Expanded scaffold flow: added step-by-step inference, proposal, user confirmation, and automated dashboard build.
- Clarified environment variable use, platform compatibility, and installation steps.
- Improved documentation for easier onboarding and maintenance.
元数据
常见问题
Glancely 是什么？
Personal tracker framework. Scaffold new habit/workout/mood/reminder trackers from natural language. Dashboard, cron, all in one skill. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 26 次。
如何安装 Glancely？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install glancely」即可一键安装，无需额外配置。
Glancely 是免费的吗？
是的，Glancely 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Glancely 支持哪些平台？
Glancely 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（macos, linux）。
谁开发了 Glancely？
由 JunjieBusiness（@junjieyu95）开发并维护，当前版本 v0.2.0。
推荐 Skills