← Back to Skills Marketplace
Qwen Qwen3
by
Twin Geeks
· GitHub ↗
· v1.0.2
· MIT-0
134
Downloads
1
Stars
2
Active Installs
3
Versions
Install in OpenClaw
/install qwen-qwen3
Description
Qwen Qwen3 — run Qwen3.5, Qwen3, Qwen3-Coder, Qwen2.5-Coder, and Qwen3-ASR across your local fleet. LLM inference, code generation, and speech-to-text from A...
Usage Guidance
Before installing: (1) verify the provenance of the pip package (check the PyPI page and the linked GitHub repo and confirm the publisher), because 'pip install' runs code on your machine. (2) Ensure you have the ollama CLI installed from its official source — the SKILL.md uses 'ollama' but the metadata doesn't declare it. (3) The instructions also reference 'uv' for ASR installs; confirm that tool or the intended installer. (4) Expect large disk, memory, and network usage when pulling models; run in an isolated VM/container if you want to reduce risk. (5) Check ~/.fleet-manager files the service will create and ensure firewall/network exposure is acceptable (services listen on localhost:11435 by default). If you are unsure about the pip package or binary provenance, do not install it on sensitive systems; instead review the package source code or run it in an isolated environment.
Capability Assessment
Purpose & Capability
The SKILL.md's purpose (run Qwen models via an Ollama-based fleet router) matches the commands and examples. However, the declared required binaries (curl or wget, optional python/pip) omit other tools the instructions actually rely on: 'ollama' is used for pulling models and calling the Ollama API, and 'uv' (used for mlx-qwen3-asr install) is referenced but not declared. This mismatch is a coherence issue: either the skill assumes these will be preinstalled or the metadata is incomplete.
Instruction Scope
Instructions direct the agent to pip install 'ollama-herd', run 'herd' and 'herd-node', pull large models via 'ollama pull', and enable transcription via local API calls to localhost:11435. These actions are within the described scope (setting up a local model fleet). The instructions will create and use files under ~/.fleet-manager (latency.db, logs/herd.jsonl) and start services that listen on a local port; they do not ask for unrelated system files or external credentials.
Install Mechanism
This is instruction-only (no install spec). The SKILL.md tells users to 'pip install ollama-herd' and to run 'ollama pull' (and 'uv tool install' for ASR). Installing from PyPI and pulling models via the ollama CLI are normal, but pip installs execute arbitrary code and present supply-chain risk. No raw download URLs or archive extracts appear in the skill itself, but the skill relies on external package managers & the ollama toolchain whose provenance the user should verify.
Credentials
No environment variables or secrets are requested. The only declared config paths (~/.fleet-manager/latency.db and logs/herd.jsonl) are consistent with a local fleet manager. There is no request for unrelated credentials or wide-scoped secrets.
Persistence & Privilege
The skill is not always-enabled and does not request elevated platform privileges. It instructs installing and running local services and writing to its own config paths, which is expected for this functionality. It does not modify other skills or global agent settings in the provided instructions.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install qwen-qwen3 - After installation, invoke the skill by name or use
/qwen-qwen3 - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
Cross-platform support: macOS, Linux, and Windows. Updated OS metadata, descriptions, and hardware recommendations.
v1.0.1
- Skill renamed from "qwen" to "qwen-qwen3" for clarity and alignment.
- Description updated to emphasize Qwen3 and Qwen3-ASR model support.
- No changes to functionality or usage instructions.
- All documentation and tables remain unchanged except for the skill name and description.
v1.0.0
Initial release of qwen-fleet 1.0.0:
- Run Qwen models (Qwen3.5, Qwen3, Qwen3/2.5-Coder, Qwen ASR) locally and route requests across multiple devices via Ollama Herd.
- Supports LLM inference, code generation, and speech-to-text from Alibaba's Qwen family with zero cloud costs.
- Provides setup instructions, usage examples (OpenAI SDK, Ollama API, transcription), and hardware recommendations.
- Highlights benefits of MoE architecture, local privacy, cost savings, and integrated fleet management.
- Includes dashboard access and guardrails for safe model and file handling.
Metadata
Frequently Asked Questions
What is Qwen Qwen3?
Qwen Qwen3 — run Qwen3.5, Qwen3, Qwen3-Coder, Qwen2.5-Coder, and Qwen3-ASR across your local fleet. LLM inference, code generation, and speech-to-text from A... It is an AI Agent Skill for Claude Code / OpenClaw, with 134 downloads so far.
How do I install Qwen Qwen3?
Run "/install qwen-qwen3" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Qwen Qwen3 free?
Yes, Qwen Qwen3 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Qwen Qwen3 support?
Qwen Qwen3 is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux, windows).
Who created Qwen Qwen3?
It is built and maintained by Twin Geeks (@twinsgeeks); the current version is v1.0.2.
More Skills