← Back to Skills Marketplace
ModelPool (Free)
by
captainlee
· GitHub ↗
· v1.0.3
· MIT-0
374
Downloads
0
Stars
3
Active Installs
3
Versions
Install in OpenClaw
/install modelpool-free
Description
ModelPool (Free) — Free AI model manager for OpenClaw. One command to auto-discover free models, configure multi-key rotation for multiplied free quota, buil...
Usage Guidance
This package appears to do what it says (manage OpenRouter keys and configure OpenClaw), but it makes significant local changes: it saves API keys in plaintext under ~/.openclaw/modelpool_keys.json, backs up and overwrites your OpenClaw config (~/.openclaw/openclaw.json), and restarts the OpenClaw gateway. Before installing or running: 1) Inspect the shipped scripts (freeswitch.py and repair.py) yourself and confirm you trust their source (registry shows unknown source/homepage). 2) Back up ~/.openclaw/openclaw.json and any important data. 3) Consider running the tool in an isolated environment or container if you don't want it to restart services on your main system. 4) If you must store API keys, prefer using a secrets manager or ensure the keys file has restrictive permissions (chmod 600) and rotate/revoke keys if you suspect compromise. 5) Note the README encourages creating multiple OpenRouter accounts to multiply free quota — that may violate OpenRouter terms of service; consider provider policy before following that guidance. If you want higher assurance, ask the publisher for a verified homepage/repo or run the tool in a disposable VM to observe behavior first.
Capability Analysis
Type: OpenClaw Skill
Name: modelpool-free
Version: 1.0.3
The ModelPool bundle is a utility for managing free AI models from OpenRouter within the OpenClaw environment. The core logic in `freeswitch.py` and `repair.py` handles API key validation, model discovery, quality ranking, and automated configuration of the OpenClaw provider settings. It includes a diagnostic tool that checks system resources (memory, disk) and connectivity. The code uses Python's standard library to avoid external dependencies and performs legitimate configuration management and service restarts without evidence of data exfiltration, obfuscation, or malicious intent.
Capability Assessment
Purpose & Capability
Name/description (multi-key rotation, model discovery, repair) match the code and README. The package only talks to OpenRouter APIs and manipulates OpenClaw configuration and services — all expected for a model manager.
Instruction Scope
SKILL.md instructs users to run interactive setup which will prompt for OpenRouter API keys. The shipped scripts make network calls only to openrouter.ai, read/write OpenClaw config files (~/.openclaw/openclaw.json, cache, and keys file), and run local OpenClaw commands. That is within scope, but the skill will store keys and change/backup config files and restart the gateway — these are material side-effects the user should expect and approve.
Install Mechanism
No install spec in the registry, but the package contains a normal setup.py and console entry point (modelpool -> freeswitch:main). Installation via pip or running the script will execute the provided Python code. Absence of a declared install mechanism in the registry means you should be careful to inspect the distributed files before running pip install or the script.
Credentials
The skill requests no environment variables but requires users to input OpenRouter API keys. The keys are saved plaintext to ~/.openclaw/modelpool_keys.json and used to call openrouter.ai — this is functional but sensitive. There are no unrelated credentials requested.
Persistence & Privilege
The skill writes to and backs up OpenClaw config, creates cache and keys files under the user's home directory, removes old logs under /tmp/openclaw, and restarts the OpenClaw daemon. It does not request always:true and does not modify other skills' configs, but these actions are privileged for the OpenClaw runtime and warrant user review.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install modelpool-free - After installation, invoke the skill by name or use
/modelpool-free - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
Fix suspicious download indicators: replaced placeholder GitHub URLs, corrected script path in README, regenerated package metadata; no functional code changes.
v1.0.2
- Added _meta.json for improved metadata handling.
- Updated several scripts (freeswitch.py, repair.py) with refinements and potential bug fixes.
- Removed the LICENSE file (license still indicated as MIT in SKILL.md).
- Updated SKILL.md with improved documentation and included prior security fixes summary.
- General maintenance and cleanup in setup files for improved stability.
v1.0.0
Initial release: auto-discover free models, multi-key rotation, smart fallback, one-click repair
Metadata
Frequently Asked Questions
What is ModelPool (Free)?
ModelPool (Free) — Free AI model manager for OpenClaw. One command to auto-discover free models, configure multi-key rotation for multiplied free quota, buil... It is an AI Agent Skill for Claude Code / OpenClaw, with 374 downloads so far.
How do I install ModelPool (Free)?
Run "/install modelpool-free" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is ModelPool (Free) free?
Yes, ModelPool (Free) is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does ModelPool (Free) support?
ModelPool (Free) is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created ModelPool (Free)?
It is built and maintained by captainlee (@meilihulee); the current version is v1.0.3.
More Skills