RHClaw (RunningHub) skill
RHClaw
Images, video, audio and 3D for OpenClaw and DeepSeek Harness: 420 RunningHub endpoints and ComfyUI apps run from a link.
- 148 ★ on the repository
- Last commit
- Apache-2.0
Extended access
Why this level
- Each run spends money from the RunningHub balance, and prompts and files are sent to an external service.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/rhclawDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Each run spends money from the RunningHub balance, and prompts and files are sent to an external service.
This is third-party code. Review the repository files before installing.
What it does
The `runninghub` skill connects OpenClaw and DeepSeek Harness to the RunningHub API. The catalog has 420 standard endpoints: 102 for images, 230 for video, 20 for audio, 16 for 3D and 52 for understanding text, images and video. A separate client runs RunningHub AI apps, which are user-made ComfyUI workflows. Before generating, the agent shows a short list of models to choose from, for example Google Veo 3.1 Fast, Kling v3.0 Pro, Midjourney v8 and GPT Image 2. The `runninghub.py` and `runninghub_app.py` scripts check the key and balance, list endpoints and run tasks.
Who it is for. OpenClaw and DeepSeek Harness users who need media generation without wiring up each API separately.
Good fit when
- An agent must draw a picture and then turn it into a video.
- A photo has to be upscaled to 4K.
- Voiceover, music or voice cloning is needed.
- A ready ComfyUI workflow from RunningHub is run from an app link.
Not a fit when
- Generation must stay local: requests go to the RunningHub cloud.
- The balance is empty: API calls are paid.
- The agent is neither OpenClaw nor DeepSeek Harness: the install is described for them.
Example request
Draw a puppy in a park, then make a five-second video from the picture, showing the model list first.Limitations
A RunningHub API key and a positive balance are required: each call spends money. The install is described as a request to the agent in an OpenClaw or DeepSeek Harness chat: it clones the repository and copies files into the workspace. Requests and uploaded files go to RunningHub servers, so private material is better kept out. The model list and discounts are set by the authors and may go stale; the endpoint catalog is refreshed with `build_capabilities.py`. Documentation is in Chinese, with an English version in `README_en.md`. The last commit in the repository dates from August 2026.
How to disable. Delete the `runninghub` folder from the agent workspace and revoke the API key on the API management page of the RunningHub account.
FAQ
How do I check the key and balance?
`python3 scripts/runninghub.py --check` verifies the key and shows the balance.
How do I list endpoints?
`--list` with `--type` and `--task` filters prints the catalog, and `--info ENDPOINT` shows the parameters of one endpoint.
How do I run an AI app?
`runninghub_app.py --info WEBAPP_ID` shows editable nodes, and `--run WEBAPP_ID` runs the app with given parameters and files.
Which model is the default?
Google Veo 3.1 Fast for video and Nano Banana Pro for images, unless another one is picked from the list.
How do I update the skill?
Ask the agent to update the skill from the repository and reread `runninghub/SKILL.md`. The key does not need to be entered again.
Discussion
No questions or reviews yet.
Related
SkillsA skill and CLI that turn saved ComfyUI workflows into agent commands, with a parameter schema and several servers
A skill and MCP driver that let Claude Code, Codex and Gemini CLI run a local ComfyUI, with prompt recipes for 68 models
Local video with MiniMax H3 on your own GPU: the open-video CLI and a skill that lets an agent write prompts and run clips.
RunAPI MCP server: a catalog of 160 models for images, video, music and speech, pricing, task runs and balance
A skill that searches the Banodoco community corpus for video and image generation tips on Wan, LTX and ComfyUI.
A Comfy Cloud plugin for Claude Code: commands to generate images, video, audio and 3D, connected through a hosted MCP server
README in short
The agent picks a fitting endpoint for the request. For AI apps it reads node information, helps set parameters and runs the app, and also lists recommended, popular and new ones.