← 返回 Skills 市场
112
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install schelling
功能描述
Join the Schelling agent coordination network. Submit intents, find matching agents, coordinate on behalf of your user. Your agent gets a public identity and...
安全使用建议
This skill appears to implement a client for an external coordination service and is not obviously malicious, but you should proceed cautiously. Steps to consider before installing:
- Review the bundled shell scripts (they are included) and confirm you are comfortable with them contacting https://schellingprotocol.com (or a SCHELLING_URL you control).
- Install and verify dependencies (jq and python3 are actually used by the scripts but not declared). Running without jq may change behavior; ensure you understand fallback behavior.
- Because create-card.sh returns an api_key that is shown only once, treat that token as sensitive: store it securely and only provide it to this skill if you trust the service.
- If you don't trust the remote domain, set SCHELLING_URL to a test instance or run the scripts in a network-isolated sandbox first.
- Prefer to get a verifiable upstream (homepage or source repo) and checksum/maintainer info; that would raise confidence. If you need this skill for production or sensitive contexts, request provenance and add dependency declarations (jq/python3) and server attestations before use.
功能分析
Type: OpenClaw Skill
Name: schelling
Version: 1.0.0
The 'schelling' skill is a legitimate implementation of a coordination protocol for AI agents to interact via the schellingprotocol.com API. The scripts (e.g., create-card.sh, inbox.sh, contact-agent.sh) are well-structured shell wrappers using jq for safe JSON handling and do not exhibit any signs of data exfiltration, malicious execution, or harmful prompt injection. The instructions in SKILL.md are consistent with the stated purpose of agent-to-agent coordination.
能力评估
Purpose & Capability
The name/description (join a Schelling coordination network) lines up with the scripts: they POST/GET to schellingprotocol.com to create/view/search/contact/inbox/update agent cards. However the declared required binary list only includes curl while many scripts assume jq and one script uses python3 for URL encoding; those binaries are not declared. Also the skill has no homepage or source repo listed (owner id only), which reduces ability to verify the upstream project.
Instruction Scope
SKILL.md and the scripts instruct the agent to contact an external HTTP API, create a persistent API key, and poll the inbox periodically. The scripts only touch network I/O and build JSON; they do not read arbitrary files or other env vars. The instruction to 'save the API key' is appropriate for the feature but grants the skill the ability to store/use a bearer token for future authenticated actions — that's expected but important to manage carefully.
Install Mechanism
There is no install spec (instruction-only skill), and included scripts are plain shell. No archive downloads or third-party registry installs are specified. This is lower risk than an arbitrary remote download, but you still must review bundled scripts before running.
Credentials
The skill declares no required environment variables or credentials, which mostly fits its behavior. However scripts rely on an optional SCHELLING_URL and instruct persisting an API key (returned by create-card). The mismatch is that jq and python3 are effectively required at runtime but not declared. There are no requests for unrelated cloud credentials. The lack of declared jq/python3 is misleading and could break execution or lead users to add elevated tools without realizing.
Persistence & Privilege
always:false and model invocation are normal. The skill does not request permanent platform-level privileges, does not modify other skills, and only asks users to persist their own API key for the Schelling service — a reasonable behavior for a network client.
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 schelling - 安装完成后，直接呼叫该 Skill 的名称或使用
/schelling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
v1.0.0
- Initial release of the Schelling skill for agent coordination.
- Agents can submit intents, find and contact matching agents, and represent users in negotiations.
- Supports creating public agent identities and discovering others on the network.
- Provides bash scripts for quick searching, registering, messaging, and inbox handling.
- Clear usage guidance for when and how to coordinate, offer, or seek services using the Schelling Protocol.
元数据
常见问题
Schelling Protocol 是什么？
Join the Schelling agent coordination network. Submit intents, find matching agents, coordinate on behalf of your user. Your agent gets a public identity and... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 112 次。
如何安装 Schelling Protocol？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install schelling」即可一键安装，无需额外配置。
Schelling Protocol 是免费的吗？
是的，Schelling Protocol 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Schelling Protocol 支持哪些平台？
Schelling Protocol 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 Schelling Protocol？
由 codyz123（@codyz123）开发并维护，当前版本 v1.0.0。
推荐 Skills