← 返回 Skills 市场
93
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install monitor-openclaw
功能描述
OpenClaw-optimized skill for Texas residential electricity shopping, address completion, candidate confirmation, ESIID lookup, usage estimation, plan recomme...
安全使用建议
This skill appears to implement the advertised Texas electricity address and plan workflows, but it has a hidden requirement: the bundled Python client expects POWERLEGO_API_TOKEN in the environment even though the skill metadata lists no required credentials. Before installing or enabling this skill: (1) verify you trust the POWERLEGO API owner (powerlego.com) and Personalized Energy (personalized.energy); (2) require the publisher to declare the POWERLEGO_API_TOKEN requirement in the skill metadata and explain the token's minimum privileges; (3) understand that user addresses (PII) will be sent to external services for validation and usage/plan lookup; (4) run the skill in a sandboxed environment or with a limited-scoped/test API token first; and (5) ask the developer to surface failure modes (what happens if the token is missing) and to document privacy/retention details. The current mismatch between code and declared requirements is the primary reason this is rated 'suspicious.'
功能分析
Type: OpenClaw Skill
Name: monitor-openclaw
Version: 1.0.1
The skill bundle is a legitimate tool designed for Texas residential electricity plan comparison and monitoring. It utilizes a series of Python scripts (e.g., scripts/fetch_best_plan.py, scripts/powerlego_api.py) to normalize user-provided addresses and interact with the PowerLego API (powerlego.com) to retrieve utility data and energy plans. The instructions in SKILL.md and system_prompt.txt are focused on maintaining a professional advisor persona and preventing the AI agent from leaking technical implementation details or internal diagnostics, which is a standard UX and safety practice. No evidence of malicious intent, unauthorized data exfiltration, or harmful execution was found.
能力评估
Purpose & Capability
The skill's name, description, and scripts align with Texas address normalization, ESIID/usage estimation, and plan lookup. However, the included Python client (scripts/powerlego_api.py) requires a POWERLEGO_API_TOKEN environment variable to call powerlego.com endpoints, yet the skill metadata declares no required environment variables or primary credential — this is inconsistent and unexplained.
Instruction Scope
The SKILL.md instructs the agent to execute local Python scripts that perform network calls to third-party APIs (powerlego.com and personalized.energy). That behavior is coherent with the purpose, but SKILL.md and the user-facing guardrails explicitly forbid revealing internal APIs/tokens, so the runtime will call external services while not exposing or documenting those calls to the user. Also, the runtime instructions do not mention the need for an API token, so running the scripts will fail (or silently error) without that secret.
Install Mechanism
There is no install spec (instruction-only style) and code files are bundled. No external binary downloads or extraction steps are specified. Risk is moderate because bundled executable Python scripts will be run, but there is no opaque installer URL or third-party package pulls.
Credentials
The code reads a sensitive environment variable named POWERLEGO_API_TOKEN (and enforces its presence). The skill metadata did not declare any required env vars or a primary credential. Asking for a single service API token would be proportionate for plan lookups, but the omission in metadata is a transparency problem. Additionally, the scripts will transmit user-supplied addresses (PII) to external endpoints (powerlego.com and personalized.energy), so the user should expect address data to leave their environment.
Persistence & Privilege
The skill is not marked always:true, does not request persistence or system-wide configuration changes, and does not declare writes to other skills' configs. It behaves as a normal user-invocable skill.
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 monitor-openclaw - 安装完成后，直接呼叫该 Skill 的名称或使用
/monitor-openclaw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
v1.0.1
- Refined operating style and user-facing voice: now enforces a strict, advisor-led experience for Texas electricity workflows, with no technical or backend language exposed.
- Implements state-machine order: step-by-step intent recognition, address normalization, candidate confirmation, and downstream plan or monitoring actions.
- Requires exact candidate confirmation before plan recommendations or URL generation.
- New rule sets for user interactions: only concise, concrete next steps and one follow-up for missing info.
- Strengthened data contract: mandates use of normalized script outputs; forbids showing raw APIs or diagnostics to users.
- Use this variant when strict workflow branching and fixed response shapes are required.
元数据
常见问题
Monitor 是什么？
OpenClaw-optimized skill for Texas residential electricity shopping, address completion, candidate confirmation, ESIID lookup, usage estimation, plan recomme... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 93 次。
如何安装 Monitor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install monitor-openclaw」即可一键安装，无需额外配置。
Monitor 是免费的吗？
是的，Monitor 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Monitor 支持哪些平台？
Monitor 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 Monitor？
由 catkennel（@catkennel）开发并维护，当前版本 v1.0.1。
推荐 Skills