← 返回 Skills 市场
Model Routing Orchestrator
作者
stephensu66
· GitHub ↗
· v2026.3.27
· MIT-0
192
总下载
3
收藏
1
当前安装
2
版本数
在 OpenClaw 中安装
/install model-routing-orchestrator
功能描述
Route each user request to the most cost-effective model or multi-model workflow based on task type, complexity, risk, latency, budget, tool needs, and verif...
安全使用建议
This skill appears coherent for deciding which model or workflow to use and does not request credentials or install code. Before enabling it in production, confirm what your deployment supplies as 'system context' (do not expose secrets, full logs, or unrelated environment variables). Test routing decisions in a sandbox with limited agent privileges, and add explicit constraints or a policy that prevents the agent from reading sensitive files or env vars when applying the routing rules. If you plan to connect provider-specific models, prefer making those provider credentials explicit in the deployment (so the skill cannot implicitly access broad environment state).
功能分析
Type: OpenClaw Skill
Name: model-routing-orchestrator
Version: 2026.3.27
The skill bundle is a comprehensive set of instructions for an AI agent to act as a model routing orchestrator, optimizing for cost and task quality. It contains no executable code, network calls, or data exfiltration logic, and the instructions in SKILL.md are strictly aligned with its stated purpose of categorizing tasks and selecting appropriate model tiers or workflows.
能力评估
Purpose & Capability
The name/description (model routing/orchestration) matches the SKILL.md content: heuristics for choosing model tiers, workflows, and verification strategies. There are no unrelated env vars, binaries, or install steps requested that would be inconsistent with routing decisions.
Instruction Scope
The instructions enumerate inputs to collect from the request and the 'system context' and give a detailed taxonomy and routing rules. However, 'system context' is open-ended and could be interpreted to include logs, environment variables, agent state, or other data sources not strictly needed for routing. The SKILL.md does not explicitly instruct reading secrets or files, but lacks explicit bounds on what counts as system context.
Install Mechanism
No install specification or code files; this is instruction-only, so nothing will be written to disk or downloaded at install time.
Credentials
The skill declares no required environment variables, credentials, or config paths. That matches the described functionality (routing heuristics) and is proportionate. Be aware the runtime wording could prompt the agent to consult environment data if available.
Persistence & Privilege
always is false and it uses normal autonomous invocation defaults. The skill does not request elevated or permanent presence or modification of other skills/configs.
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 model-routing-orchestrator - 安装完成后，直接呼叫该 Skill 的名称或使用
/model-routing-orchestrator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
v2026.3.27
No user-visible changes. Version bumped from 2026.3.26 to 2026.3.27 with no file or documentation updates.
v2026.3.26
Initial commit.
元数据
常见问题
Model Routing Orchestrator 是什么？
Route each user request to the most cost-effective model or multi-model workflow based on task type, complexity, risk, latency, budget, tool needs, and verif... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 192 次。
如何安装 Model Routing Orchestrator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install model-routing-orchestrator」即可一键安装，无需额外配置。
Model Routing Orchestrator 是免费的吗？
是的，Model Routing Orchestrator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Model Routing Orchestrator 支持哪些平台？
Model Routing Orchestrator 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 Model Routing Orchestrator？
由 stephensu66（@stephensu66）开发并维护，当前版本 v2026.3.27。
推荐 Skills