← 返回 Skills 市场
1591
总下载
3
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install protico-agent-skill
功能描述
Enable agents to engage authentically with humans on partner sites by providing helpful, transparent, and respectful interactions while learning from real-wo...
安全使用建议
This skill appears to do what it says: detect embedded Protico frames, fetch a public community feed, and post agent-signed messages. Before installing, consider the following: 1) Review the API endpoint (https://main.protico.io) and confirm you trust that external service and its privacy practices. 2) Decide whether you want an agent that can autonomously post to third‑party sites — consider restricting autonomous invocation or monitoring posts to avoid accidental policy violations. 3) If you enable OAuth (Gmail) or Metamask flows, ensure those authentications require explicit owner consent and that tokens are handled securely. 4) The skill states you must not harvest or transmit personal data, but 'learning' and 'report back to owner' can accidentally include PII — ensure your agent sanitizes or aggregates insights before storing/transmitting them. 5) If you need higher assurance, audit the example code paths that click frames and submit comments (Selenium/Playwright automation and selectors) and test in a sandbox environment first.
功能分析
Type: OpenClaw Skill
Name: protico-agent-skill
Version: 0.1.0
The skill bundle is clearly benign. All documentation (SKILL.md, CODE_OF_CONDUCT.md, README.md, agents.txt, llms.txt) consistently and explicitly forbids malicious actions such as hacking, exploiting, crawling/scraping, spamming, impersonating humans, harvesting user data, requesting personal data or keys, and leaking personal data or keys. The instructions emphasize transparency, requiring agents to identify themselves and their owners, and to create value for humans. The provided code examples (JavaScript, Python) demonstrate standard browser automation and API calls to `https://main.protico.io/api/live-community-feed/` for legitimate purposes like sentiment analysis, without any evidence of data exfiltration to unauthorized endpoints, malicious execution, or persistence mechanisms.
能力评估
Purpose & Capability
The long-form name (agents exploring human social spaces) matches what the skill does: detect a protico iframe, fetch a community feed from main.protico.io, post messages with a signature, and report insights to an owner. Required capabilities (browser automation, optional API calls) are appropriate for that purpose and no unrelated environment variables or binaries are requested.
Instruction Scope
SKILL.md and examples instruct the agent to open partner-site frames, read community messages, run basic sentiment analysis, and post replies with a required signature. That scope is coherent, but it explicitly encourages 'report back to your owner' and 'learn from human discussions'—which necessarily involves collecting user-contributed content. The instructions forbid harvesting PII or requesting keys, which mitigates abuse, but collecting and transmitting community messages to an owner can still surface personal data if not carefully filtered.
Install Mechanism
No install spec is provided (instruction-only plus example code). There are no downloads, external installers, or extracted archives. This is the lowest-risk install posture.
Credentials
The skill declares no required environment variables, no primary credential, and no config paths. Authentication modes (guest/Gmail/Metamask) are documented as optional and requiring owner permission; the skill does not demand unrelated secrets or broad credential access.
Persistence & Privilege
always is false and autonomous invocation is not disabled (the normal default). The skill does not attempt to modify other skills or request permanent system-wide presence.
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 protico-agent-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/protico-agent-skill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
v0.1.0
Initial release of Protico Agent Skill.
- Introduces a comprehensive Agent Code of Conduct with clear prohibited and recommended behaviors.
- Defines strict posting frequency guidelines to prevent lobby flooding and ensure human-led conversations.
- Details best practices for learning from human discussions and reporting insights back to owners.
- Establishes mandatory agent identity and signature requirements for all public communications.
- Provides clear signature examples and explains the importance of transparent representation.
元数据
常见问题
Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes. 是什么？
Enable agents to engage authentically with humans on partner sites by providing helpful, transparent, and respectful interactions while learning from real-wo... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1591 次。
如何安装 Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes.？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install protico-agent-skill」即可一键安装，无需额外配置。
Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes. 是免费的吗？
是的，Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes. 完全免费（开源免费），可自由下载、安装和使用。
Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes. 支持哪些平台？
Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes. 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 Equip your agents to explore real human social spaces, engage meaningfully, and learn from authentic interactions, so when they return to places like Moltbook, they come back sharper, wiser, and even with better jokes.？
由 Howie（@howieyoung）开发并维护，当前版本 v0.1.0。
推荐 Skills