nadirclaw

// Install, configure, and run NadirClaw LLM router to cut AI API costs by 40-70%. Use when the user wants to reduce LLM spending, route prompts to cheaper models, set up cost-saving proxy, or optimize API usage across providers (OpenAI, Anthropic, Google, Ollama). Also use when asked about model routi

$ git log --oneline --stat

stars:1,933

forks:367

updated:March 4, 2026

SKILL.mdreadonly

SKILL.md Frontmatter

namenadirclaw

version1.0.0

descriptionInstall, configure, and run NadirClaw LLM router to cut AI API costs by 40-70%. Use when the user wants to reduce LLM spending, route prompts to cheaper models, set up cost-saving proxy, or optimize API usage across providers (OpenAI, Anthropic, Google, Ollama). Also use when asked about model routing, LLM cost optimization, or setting up NadirClaw with OpenClaw.

NadirClaw Skill

NadirClaw is an open-source LLM router that classifies prompts in ~10ms and routes simple ones to cheap/local models while keeping complex work on premium models.

Install

pip install nadirclaw

Setup

Run the interactive wizard:

nadirclaw setup

Or auto-configure for OpenClaw:

nadirclaw openclaw onboard

This writes NadirClaw as a provider in OpenClaw config with model nadirclaw/auto. No restart needed.

Start

nadirclaw serve --verbose

Runs on http://localhost:8856. Any OpenAI-compatible tool can use it by pointing to this URL.

Point tools at NadirClaw

# OpenClaw (auto)
nadirclaw openclaw onboard

# Claude Code
ANTHROPIC_BASE_URL=http://localhost:8856/v1 claude

# Any OpenAI-compatible tool
OPENAI_BASE_URL=http://localhost:8856/v1 <tool>

Routing profiles

Pass x-routing-profile header or use these models:

nadirclaw/auto - smart routing (default)
nadirclaw/eco - maximize savings
nadirclaw/premium - always use best model
nadirclaw/free - Ollama/local only
nadirclaw/reasoning - chain-of-thought optimized

Monitor savings

nadirclaw savings      # cost savings report
nadirclaw report       # detailed routing analytics
nadirclaw dashboard    # live terminal dashboard

Key features

~10ms classification overhead
Session persistence (no model bouncing mid-conversation)
Rate limit fallback (auto-retry on 429)
Agentic task detection (forces premium for tool use)
Context-window filtering (auto-swaps for long conversations)
Supports: OpenAI, Anthropic, Google Gemini, Ollama, any LiteLLM provider

Troubleshooting

If nadirclaw serve fails, check API keys: nadirclaw setup
For Ollama: ensure ollama serve is running first
Logs: nadirclaw report --last 20 to see recent routing decisions
Raw debug: nadirclaw serve --verbose --log-raw