Free Ride 1.0.4

Manages free AI models from OpenRouter for OpenClaw. Automatically ranks models by quality, configures fallbacks for rate-limit handling, and updates opencla...

Security Scan

The skill's code and instructions largely match its stated purpose (managing OpenRouter free models) but there are several metadata and packaging inconsistencies you should understand before installing.

Purpose & Capability

The skill's name/description (manage OpenRouter free models) aligns with the included Python code that calls openrouter.ai and updates ~/.openclaw/openclaw.json. However the registry/header metadata published with this evaluation is inconsistent: the top-level 'Requirements' block said 'Required env vars: none' while SKILL.md and skill.json clearly require OPENROUTER_API_KEY. The skill also contains install packaging (setup.py, entry points) despite the registry saying 'No install spec — this is an instruction-only skill.' These metadata mismatches reduce trust and should be resolved before install.

Instruction Scope

SKILL.md instructs the agent/user to set OPENROUTER_API_KEY, install the local freeride package (pip install -e . in ~/.openclaw/workspace/skills/free-ride), run freeride commands, and restart the OpenClaw gateway. Those actions are within the stated purpose (fetching models, ranking them, and writing OpenClaw config). The instructions ask the agent to read and write the user's OpenClaw config (~/.openclaw/openclaw.json) and to run the local CLI — expected for this functionality. Note: SKILL.md advises using openclaw config set env... which would store the API key in OpenClaw config; ensure you accept that storage location.

Install Mechanism

There is no remote binary download; installation is a local pip install of the packaged code (setup.py). That's a low-to-moderate risk pattern but the registry text earlier incorrectly said 'instruction-only'. The skill.json includes an 'install' recommendation (npx clawhub... && pip install -e .). No external archives or downloads from untrusted URLs are used in the code. Still, pip installing local code executes Python code on your machine — review files before installing.

Credentials

The only credential the code actually needs is an OpenRouter API key (OPENROUTER_API_KEY), which is proportionate to the described purpose. However, the manifest presented to you earlier claimed no required env vars while skill.json and SKILL.md require OPENROUTER_API_KEY. That mismatch should be corrected. The code reads the key from the environment or from ~/.openclaw/openclaw.json; it does not appear to request unrelated credentials.

Persistence & Privilege

The skill writes cache and state files under ~/.openclaw (cache, .freeride-watcher-state.json) and updates ~/.openclaw/openclaw.json to set primary and fallback models. 'always' is false and autonomous invocation is allowed (default). Writing the OpenClaw config and state files is expected for this feature and is scoped to the user's OpenClaw directory; it does not request system-wide or other skills' credentials.

What to consider before installing

This skill's code appears to implement what it claims (fetching free models from OpenRouter, ranking them, and updating your OpenClaw config), but there are a few red flags to check before installing: - Confirm the OPENROUTER_API_KEY requirement: SKILL.md and skill.json require it despite header metadata saying 'none'. Only provide an API key you trust and prefer one scoped/limited if possible. The code reads the key from your environment or from ~/.openclaw/openclaw.json. - Inspect the packaged files (main.py, watcher.py, setup.py, README, skill.json) locally before running pip install -e . — installing executes local code on your machine. - Backup ~/.openclaw/openclaw.json before running FreeRide so you can restore configuration if changes are unwanted. - The watcher daemon will make outbound calls to openrouter.ai and will write state and cache files under ~/.openclaw; if you want to avoid background activity, run only the CLI commands or run the watcher in a controlled environment (or as a one-shot cron job) rather than a continuous daemon. - Verify the skill's provenance: the manifest includes differing owner IDs and the package's 'install' path references the local workspace. If you care about supply-chain trust, find the GitHub repository (skill.json points to a repo) and verify the source and publisher before trusting and installing. Given the inconsistencies in metadata and packaging, proceed with caution: reasonable and expected behavior is present (verdict not 'benign'), but the metadata mismatches and local installation requirement mean you should manually review files and back up config before installing.

FreeRide - Free AI for OpenClaw

What This Skill Does

Configures OpenClaw to use free AI models from OpenRouter. Sets the best free model as primary, adds ranked fallbacks so rate limits don't interrupt the user, and preserves existing config.

Prerequisites

Before running any FreeRide command, ensure:

1. OPENROUTER_API_KEY is set. Check with echo $OPENROUTER_API_KEY. If empty, the user must get a free key at https://openrouter.ai/keys and set it:

   export OPENROUTER_API_KEY="sk-or-v1-..."
   # Or persist it:
   openclaw config set env.OPENROUTER_API_KEY "sk-or-v1-..."
   

2. The freeride CLI is installed. Check with which freeride. If not found:

   cd ~/.openclaw/workspace/skills/free-ride
   pip install -e .
   

Primary Workflow

When the user wants free AI, run these steps in order:

# Step 1: Configure best free model + fallbacks
freeride auto

# Step 2: Restart gateway so OpenClaw picks up the changes
openclaw gateway restart

That's it. The user now has free AI with automatic fallback switching.

Verify by telling the user to send /status to check the active model.

Commands Reference

Command	When to use it
freeride auto	User wants free AI set up (most common)
freeride auto -f	User wants fallbacks but wants to keep their current primary model
freeride auto -c 10	User wants more fallbacks (default is 5)
freeride list	User wants to see available free models
freeride list -n 30	User wants to see all free models
freeride switch <model>	User wants a specific model (e.g. freeride switch qwen3-coder)
freeride switch <model> -f	Add specific model as fallback only
freeride status	Check current FreeRide configuration
freeride fallbacks	Update only the fallback models
freeride refresh	Force refresh the cached model list

After any command that changes config, always run openclaw gateway restart.

What It Writes to Config

FreeRide updates only these keys in ~/.openclaw/openclaw.json:

• agents.defaults.model.primary — e.g. openrouter/qwen/qwen3-coder:free
• agents.defaults.model.fallbacks — e.g. ["openrouter/free", "nvidia/nemotron:free", ...]
• agents.defaults.models — allowlist so /model command shows the free models

Everything else (gateway, channels, plugins, env, customInstructions, named agents) is preserved.

The first fallback is always openrouter/free — OpenRouter's smart router that auto-picks the best available model based on the request.

Watcher (Optional)

For auto-rotation when rate limited, the user can run:

freeride-watcher --daemon    # Continuous monitoring
freeride-watcher --rotate    # Force rotate now
freeride-watcher --status    # Check rotation history

Troubleshooting

Problem	Fix
freeride: command not found	cd ~/.openclaw/workspace/skills/free-ride && pip install -e .
OPENROUTER_API_KEY not set	User needs a key from https://openrouter.ai/keys
Changes not taking effect	openclaw gateway restart then /new for fresh session
Agent shows 0 tokens	Check freeride status — primary should be openrouter/<provider>/<model>:free