← 返回 Skills 市场
ZworkerAgentSkillOpenClaw
作者
peterpan0630
· GitHub ↗
· v1.0.2
· MIT-0
76
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install zworkeragentskillopenclaw
功能描述
控制zworker AI自动化任务应用，支持任务管理、定时计划控制、用户信息同步和消息通知转发。当用户提到“zworker”关键词时使用此技能，例如：同步用户信息、获取任务列表、执行任务、管理定时计划、获取并发送通知等。技能通过HTTP接口与本地运行的zworker应用（localhost:18803）通信。
安全使用建议
What to consider before installing:
- This skill talks to a local service on http://localhost:18803 and includes client scripts that also target only localhost — that is expected for a local integrator, but ensure the zworker app running there is trusted.
- SKILL.md says the agent should read OpenClaw's openclaw.json and use an OpenClaw message tool, but the packaged scripts do not automatically read that file or send messages; the agent (or additional glue) would need file access or message tooling to perform those steps. Ask the publisher to clarify the exact file path and message-sending mechanism.
- Because the instructions imply reading local config files, verify what openclaw.json contains (sensitive tokens?) before allowing the skill to run. If openclaw.json contains secrets, do not grant autonomous invocation until you can restrict access.
- If you plan to let the agent invoke this skill autonomously, restrict it unless you trust the local zworker service and have reviewed openclaw.json and any message-sending tools it might use.
- To increase confidence: request updated SKILL.md that (a) lists explicit config file paths required, (b) documents how messages are delivered (which tool/command and whether credentials are needed), and (c) explains why no env vars/config paths are declared despite needing local files/tools.
功能分析
Type: OpenClaw Skill
Name: zworkeragentskillopenclaw
Version: 1.0.2
The skill bundle provides a legitimate integration between OpenClaw and a local AI automation application called 'zworker' running on localhost:18803. The scripts (e.g., zworker_api.py, sync_users.py) and SKILL.md instructions facilitate task management, schedule control, and notification forwarding as described. All network activity is restricted to the local loopback address, and there is no evidence of data exfiltration to external endpoints, obfuscation, or malicious intent.
能力评估
Purpose & Capability
The skill's name/description describe controlling a local zworker service on localhost:18803, which matches the included Python client and scripts. However, the SKILL.md claims the agent will read OpenClaw's openclaw.json and use an OpenClaw 'message' tool to send messages, while the manifest declares no required config paths or environment variables and the provided scripts do not implement automatic reading of openclaw.json or sending via a message tool. That mismatch (declared zero requirements but runtime expectations for local files/tools) is disproportionate and unexplained.
Instruction Scope
SKILL.md instructs reading local OpenClaw configuration (openclaw.json) and using an external 'message' tool to forward notifications. The included scripts instead expect input via stdin, CLI args, or files (sync_users.py) and only output notification data (fetch_notifications.py) rather than sending messages themselves. This gives the agent broad discretion to read local files or invoke host messaging tooling at runtime; the instructions lack concrete, constrained paths or explicit message-sending implementation.
Install Mechanism
No install specification is provided (instruction-only with included scripts). Nothing is downloaded or written by an install step. Included Python scripts are present but no package installation or network fetch occurs as part of installation.
Credentials
The skill declares no required environment variables or credentials, yet operation depends on local artifacts and tools (openclaw.json, an OpenClaw message-sending tool, and a local zworker HTTP service). Because those are not declared as required config paths or env vars, it's unclear what the agent will try to access at runtime. There is no request for remote credentials, and network access is limited to localhost in the code, which is appropriate — but the absence of declared local file or tool requirements is a mismatch.
Persistence & Privilege
Flags: always=false and disable-model-invocation=false (normal). The skill does not request permanent/system-wide presence, does not modify other skills' configuration files, and does not claim elevated privileges.
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 zworkeragentskillopenclaw - 安装完成后，直接呼叫该 Skill 的名称或使用
/zworkeragentskillopenclaw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
v1.0.2
zworkeragentskillopenclaw v1.0.2
- 新增完善的中文文档（SKILL.md），详述所有功能、接口与使用方法。
- 明确列出同步用户信息、消息通知、任务及定时计划管理等核心能力。
- 详细描述各操作的输入语句、接口调用流程和输出格式。
- 列出所有脚本工具及关键配置说明，便于快速集成和故障排查。
- 重点强调本地HTTP接口调用与后续可能的认证需求。
元数据
常见问题
ZworkerAgentSkillOpenClaw 是什么？
控制zworker AI自动化任务应用，支持任务管理、定时计划控制、用户信息同步和消息通知转发。当用户提到“zworker”关键词时使用此技能，例如：同步用户信息、获取任务列表、执行任务、管理定时计划、获取并发送通知等。技能通过HTTP接口与本地运行的zworker应用（localhost:18803）通信。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 76 次。
如何安装 ZworkerAgentSkillOpenClaw？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install zworkeragentskillopenclaw」即可一键安装，无需额外配置。
ZworkerAgentSkillOpenClaw 是免费的吗？
是的，ZworkerAgentSkillOpenClaw 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
ZworkerAgentSkillOpenClaw 支持哪些平台？
ZworkerAgentSkillOpenClaw 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 ZworkerAgentSkillOpenClaw？
由 peterpan0630（@peterpan0630）开发并维护，当前版本 v1.0.2。
推荐 Skills