ClawHub ships with a core set of skills. Most people install github and gog during setup and stop there. That is a reasonable start. But there are skills in the catalog that do genuinely useful things and get far less attention than they deserve.
This is not a sponsored post. These are skills worth knowing about because they solve real problems.
gog (Google Workspace)
Most people know gog does Gmail. What they miss is that it covers the entire Google Workspace stack.
One skill, one install, and your agent can read and search Gmail, create and update calendar events, upload and read files in Drive, read and write to Google Sheets, and create and edit Google Docs.
The Sheets integration is underrated. If you use Sheets as a lightweight database (client lists, expense tracking, content calendars), gog makes it scriptable without building a separate integration. The Expense Logger and Invoice Tracker runbooks both use gog Sheets for this reason.
Limitation: gog requires OAuth setup with your Google account. Takes 10-15 minutes the first time. Not zero-effort, but it is a one-time cost.
clawhub install gogclawhub
clawhub is the skill manager. Most people treat it as a one-time setup tool and never think about it again. That is a mistake.
Your agent can use clawhub to search for skills by name or description, install new skills when it realizes it needs a capability, update installed skills to the latest version, and list what is currently installed.
This means you can tell your agent "find a skill that lets you send WhatsApp messages" and it will search ClawHub, find the right skill, and install it. You do not need to know the exact skill name in advance.
Limitation: skills can only be installed when the agent already has the clawhub skill installed. Chicken-and-egg problem only on first setup.
clawhub install clawhub # already installed if you used the standard setuphealthcheck
healthcheck runs security audits on the machine running OpenClaw. Most people set up their agent and never think about the host security again.
What it does: checks firewall configuration, audits SSH settings, verifies automatic security updates are enabled, reports on exposed services and ports, and checks the OpenClaw version.
The output is actionable. Not "here are some security considerations" but "your SSH port is 22 and password auth is enabled, here is how to fix it."
This is particularly useful if you are running OpenClaw on a VPS or a Raspberry Pi rather than a laptop behind a home router.
Limitation: currently macOS and Debian/Ubuntu focused. Limited support for other distros.
clawhub install healthcheckvideo-frames
video-frames uses ffmpeg to extract frames or short clips from videos. Not the most common use case, but when you need it, you really need it.
Useful for: extracting a specific frame from a screen recording to use as a thumbnail, pulling frames from a video for a bug report, creating a GIF from a short clip, and batch-processing video files.
Your agent can take a video file path and a timestamp and return the frame. Or you can specify a frame rate and get a sequence of frames for analysis.
Limitation: requires ffmpeg to be installed on the host. Install it first if needed.
brew install ffmpeg # if not already installed
clawhub install video-framesweather
weather fetches current conditions and forecasts from wttr.in or Open-Meteo. No API key required. Works for any location.
The obvious use is "what is the weather today." The less obvious use is building it into other workflows: morning briefings, event planning, outdoor activity schedulers, or garden care agents that adjust watering reminders based on rain forecast.
Limitation: not suitable for severe weather alerts, historical data, or detailed meteorological analysis. It is a general-purpose forecast tool.
clawhub install weathernano-banana-pro
nano-banana-pro wraps Gemini image generation. You can generate and edit images directly from your agent without opening a separate tool.
Common uses: generating placeholder images for projects, creating thumbnails or social media images from a description, editing existing images, and rapid prototyping of visual concepts.
Limitation: image quality depends on the underlying model. Not a replacement for professional design work, but useful for quick iterations and filling gaps.
clawhub install nano-banana-proskill-creator
skill-creator is for building new skills. If you have a workflow that does not fit any existing skill, you can use skill-creator to package it properly.
A skill is a SKILL.md file plus any scripts, configs, and assets it needs. skill-creator helps you structure this correctly so the skill works with clawhub and can optionally be published to the ClawHub registry.
Useful if you have an internal tool with an API and want to give your agent access to it, you have a custom script your agent should be able to run, or you want to share a capability you built with other OpenClaw users.
Limitation: building a good skill requires some understanding of how OpenClaw skills work. Read a few existing SKILL.md files first to understand the conventions.
clawhub install skill-creatorThe install pattern
For any skill: install it, verify it installed, and ask your agent to use it. No restart required.
clawhub install <skill-name>
clawhub list # verify it installedAfter installing, ask your agent to read the SKILL.md and it will know how to use the new capability. The skills you overlook are usually the ones that save the most time once you find them.