L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

freeride

freeride

Manages free AI models from OpenRouter for OpenClaw. Automatically ranks models by quality, configures fallbacks for rate-limit handling, and updates openclaw.json. Use when the user mentions free AI, OpenRouter, model switching, rate limits, or wants to reduce AI costs.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: freeride description: Manages free AI models from OpenRouter for OpenClaw. Automatically ranks models by quality, configures fallbacks for rate-limit handling, and updates openclaw.json. Use when the user mentions free AI, OpenRouter, model switching, rate limits, or wants to reduce AI costs.

FreeRide - Free AI for OpenClaw

Configures OpenClaw to use free AI models from OpenRouter with automatic fallback switching.

Quick Start

# Set API key (free at openrouter.ai/keys)
export OPENROUTER_API_KEY="sk-or-v1-..."

# Auto-configure best model + fallbacks
freeride auto

Commands

list - View available models

freeride list              # Top 15 models
freeride list -n 30        # More models
freeride list --refresh    # Force API refresh

auto - Auto-configure

freeride auto              # Best model + 5 fallbacks
freeride auto -f           # Fallbacks only (keep current primary)
freeride auto -c 10        # 10 fallbacks
freeride auto --setup-auth # Also configure auth profile

switch - Set specific model

freeride switch qwen3-coder         # Set as primary
freeride switch deepseek -f         # Add to fallbacks only
freeride switch nvidia/nemotron --no-fallbacks

status - Check configuration

freeride status

fallbacks - Update fallbacks only

freeride fallbacks         # 5 fallbacks
freeride fallbacks -c 10   # 10 fallbacks

refresh - Update model cache

freeride refresh

Behavior

Primary model: Best specific model (not router) for consistent responses.

First fallback: Always openrouter/free - OpenRouter's smart router that auto-selects based on request features (vision, tools, etc.).

Additional fallbacks: Ranked by quality score.

Config preservation: Only updates model-related sections; preserves gateway, channels, plugins, etc.

Model Ranking

Score (0-1) based on:

  • Context length (40%)
  • Capabilities (30%)
  • Recency (20%)
  • Provider trust (10%)

Flags

FlagCommandsDescription
-fswitch, autoFallback only, keep primary
-c Nauto, fallbacksNumber of fallbacks
--no-fallbacksswitchSkip fallback configuration
--setup-authswitch, autoAdd OpenRouter auth profile
-n NlistModels to display
-rlistForce refresh

Config Output

Updates ~/.openclaw/openclaw.json:

{
  "agents": {
    "defaults": {
      "model": {
        "primary": "openrouter/qwen/qwen3-coder:free",
        "fallbacks": [
          "openrouter/free:free",
          "nvidia/nemotron-3-nano-30b-a3b:free"
        ]
      }
    }
  }
}

Troubleshooting

"OPENROUTER_API_KEY not set": Export the key or add to shell profile.

Config not updating: Check file permissions on ~/.openclaw/openclaw.json.

Changes not taking effect: Restart OpenClaw.

Skills enregistrées