vllm-projectOfficial·18 Agent Skills Included

semantic-router

Route LLM requests to the best model automatically

Routes incoming LLM requests to the right model based on content, cost, latency, and privacy rules. Removes hard-coded model selection, manual tuning, and wasted spend on oversized models. Includes signal extraction, decision policies, safety filters, and a dashboard to manage routing across mixed CPU, GPU, and cloud infrastructure.

5,474|880

↗

npx skills add vllm-project/semantic-router --all -g -y

Available:

Directs the AI agent to route each coding task to the correct sub-skill, follow the repository's validation gates, and respect local rules before editing any module.

All Skills in This Repository (18)

Pure Emerald Level Indicators

📦 In Repo

vllm-project

config-platform-change

Synchronize config concepts across router config, Python CLI schema, and dashboard UI.

signal-end-to-end

Coordinate end-to-end signal changes across router configuration, extraction, CLI schema, and metadata contracts.

dashboard-platform-change

Orchestrate dashboard frontend and backend changes to align with router contracts.

routing-policy-change

Modify post-signal routing policies for predicates, model ranking, and looper behavior.

startup-chain-change

Configure and validate the local vLLM Semantic Router startup chain.

harness-contract-change

Coordinate agent-harness contract changes across docs, manifests, and validation scripts.

cross-stack-bugfix

Diagnoses and fixes bugs spanning multiple layers with coordinated changes.

training-stack-change

Coordinate training stack changes across workflows, artifacts, and runtime outputs.

plugin-end-to-end

Coordinate plugin lifecycle changes across router config, processing, and E2E tests.

fleet-sim-change

Coordinate fleet-sim code, docs, and release workflow changes to prevent drift.

k8s-platform-change

Coordinate Kubernetes platform changes by updating operator APIs, CRDs, and deployment profiles.

router-service-platform-change

Modify router-side API, authz, memory, provider, storage, or runtime service modules with automated validation and traceability.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install vLLM Semantic Router?▼

Run `npx skills add vllm-project/semantic-router --all -g -y` in your terminal to install all skills in this suite globally.

What does vLLM Semantic Router do?▼

It sits in front of your LLM endpoints and automatically sends each request to the best model based on topic, cost, speed, and safety rules you define.

How to reduce LLM inference costs with routing?▼

The router sends simple queries to small cheap models and only escalates hard ones to expensive models, cutting spend without hurting quality.

Does it work with OpenClaw and other AI agents?▼

Yes. A built-in bridge skill imports your OpenClaw model providers and rewrites the config so your agent talks to the router automatically.

Can it detect hallucinations and jailbreaks?▼

Yes. Built-in classifiers flag hallucinated answers, jailbreak attempts, and PII in requests and responses before they reach users.