OpenClaw and Hermes skills, security-reviewed before you install them
A skill is a small package, a SKILL.md and a few scripts, that teaches an agent one job. The community publishes them to ClawHub; Diali mirrors the catalog, shows the security review on each one before you install it, and runs the skill inside your agent’s own sandbox.
Five of the most-installed skills, one page each
The pilot. Each page shows what the skill does, its security review, what it needs before the first run, and how to install it on a Diali agent.
UI/UX Pro Max
UI/UX design intelligence and implementation guidance for building polished interfaces. Use when the user asks for UI design, UX flows, information architecture, visual style direction, design systems/tokens, component specs, copy/microcopy, accessibility, or to generate/critique/refine frontend UI (HTML/CSS/JS, React, Next.js, Vue, Svelte, Tailwind). Includes workflows for (1) generating new UI layouts and styling, (2) improving existing UI/UX, (3) producing design-system tokens and component guidelines, and (4) turning UX recommendations into concrete code changes.
Humanizer
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.
Obsidian
Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude rea
What an OpenClaw skill is, in one minute
OpenClaw and Hermes both load skills the same way: a folder with a SKILL.md that tells the agent when the skill applies and how to use it, plus any scripts it needs. Installing one adds a capability, such as reading an Obsidian vault or vetting other skills, without changing the agent itself.
ClawHub is the public registry the community publishes to, and it audits listings for security. Diali mirrors that catalog into the dashboard and keeps the audit next to each skill that has one: a verdict, a plain-language summary of what the skill actually does, and the date it was checked. You read it before you install.
Once installed, a skill runs inside your agent’s own kernel-sandboxed workspace, so it can only reach what the agent can reach. Skills that need credentials or settings say so before install, and the dashboard asks for them at setup.
Questions people ask
A folder with a SKILL.md that tells the agent when the skill applies and what to do, plus any scripts it calls. OpenClaw and Hermes load skills the same way, so most skills run on both; each page here says which.
From the dashboard: open Skills, find the skill, read its security review, and install it on the agent you choose. If it needs credentials or settings, the dashboard asks for them before the first run.
Read the review first. ClawHub audits listings and Diali shows the result on each skill: a verdict, a summary of what the skill actually does, and the date it was checked. A skill with no audit says so. Whatever you install runs inside your agent’s own sandbox, not on a shared machine.
Most do. The catalog records which engines a skill supports, and every page here prints it, so you can tell before installing.
Give a skill an agent to run on
Every plan includes the skills catalog, every channel and every connector. Build an agent and install your first skill from the dashboard.