Router NIMIMORE for Openclaw

An intelligent model router that automatically selects the optimal AI model based on query complexity to balance cost and performance.

aichiafranco
v1.0.0
Mar 14, 2026
0
906
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install router-nimimore

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install router-nimimore using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Router NIMIMORE?

Router NIMIMORE is a sophisticated model orchestration tool designed for the Openclaw Skills framework. It serves as an intelligent middle layer that analyzes the intent, language, and complexity of incoming queries to assign them to the most cost-effective yet capable model. By using this skill, developers can significantly reduce API expenditures without sacrificing the quality of responses for complex reasoning tasks.

The system utilizes a tiered logic approach, ensuring that simple requests don't waste premium tokens. Whether you are handling basic chat interactions or deep architectural reasoning, Router NIMIMORE ensures your Openclaw Skills setup operates at peak efficiency by leveraging providers like Moonshot and Bailian effectively.

Router NIMIMORE Use Cases

  • Reducing API operational costs by up to 85% for simple automated tasks.
  • Automatically directing coding requests to models with high logic benchmarks.
  • Balancing context window requirements against token pricing for document summarization.
  • Enhancing agent autonomy by allowing it to self-select the best tool for a specific sub-task.

How Router NIMIMORE Works

  1. The user or agent passes a query string to the router.analyze tool.
  2. The system evaluates query characteristics such as language, required reasoning depth, and estimated context length.
  3. The router.select tool compares these requirements against the metadata of available model tiers (Premium, Standard, Economy).
  4. The optimal model endpoint is returned to the agent to execute the primary task.
  5. Developers can use router.demo to visualize the decision-making logic across different query types.

Router NIMIMORE Setup

Router NIMIMORE is designed to work out of the box with the Openclaw Skills environment. To manually trigger the router via the CLI, use the following commands:

# Route a specific query to find the best model
python scripts/router.py --query "Write a Python function for data sorting"

# Run a demonstration of the routing logic
python scripts/router.py --demo

# Select a model based on query and context size
python scripts/router.py --query "Summarize this text" --context-length 5000

Router NIMIMORE Data Schema & Taxonomy

The skill categorizes models into a structured hierarchy to facilitate precise selection within the Openclaw Skills ecosystem:

Model Tier Cost per 1k Tokens Primary Use Case Example Models
Premium $0.015 Complex reasoning & deep logic moonshot/kimi-k2.5
Standard $0.004 - $0.008 Coding and general Chinese tasks qwen-max, qwen-plus
Economy $0.002 Simple queries & fast responses qwen-turbo

Data output includes the model identifier, tier classification, and the reasoning behind the specific selection.

Router NIMIMORE Advanced Features

  • Context-aware routing that considers historical conversation length.
  • Multi-provider support across Moonshot and Alibaba Bailian platforms.
  • Automated cost-saving analysis providing up to 47% savings on standard code tasks.
  • Real-time query characteristic analysis for dynamic agent workflows.
  • Lightweight implementation requiring zero external configuration for standard Openclaw Skills deployments.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*