Awesome OpenClaw Skills: 20 ClawHub Skills Worth Installing (2026)

Looking for awesome OpenClaw skills usually means a GitHub list, a search box and a lot of guesswork. This is the ClawDocx team's own pick of 20 OpenClaw skills, each one confirmed on its ClawHub listing on 19 September 2026 and grouped by what you would actually use it for: communication, development, research, productivity, voice and media, and security. Every entry carries its owner-qualified install command and a link to the listing, so you can read the SKILL.md and ClawHub's audit status before you install anything.

#Why a Curated List?

This is the ClawDocx team's own editorial pick, grouped by use case: ClawHub publishes no top 20, no editorial ranking and no star rating for skills, so nothing below is a score, a placement or a community verdict. ClawHub does publish a Featured shelf, but for skills the docs say those sixteen slots are all telemetry selections, not editorial ones. Every entry was confirmed on ClawHub on 19 September 2026 by its owner-qualified reference, and each install command is the documented openclaw skills install @owner/<slug> form with that skill's real owner handle.

What ClawHub does publish is registry state. The ClawHub overview says the registry "tracks semver versions, tags such as latest, changelogs, files, downloads, stars, and security scan summaries", and that "Public pages show current registry state so users can inspect a skill or plugin before installing it." Stars are bookmarks rather than a rating. The ClawHub CLI page documents star and unstar in one line: "Adds/removes a skill from your Bookmarks."

No skill can award itself a badge through its topics. ClawHub does maintain a publisher official flag and a Featured marker, but neither is something a publisher sets, and neither is a review of the skill's contents. ClawHub's publishing page lists approved, audited, certified, clawhub, community, curated, endorsed, featured, official, officials, openclaw, recommended, staff-pick, trusted, trusted-publisher and verified among the topic names that "are reserved by ClawHub and are rejected", normalized before the check "so Official and staff pick are rejected too." A listing that appears to wear one of those words as a badge is not showing you a ClawHub endorsement.

#Which trust signals are real?

ClawHub's security audits page names the things worth reading before an install: "the overall audit status", "the risk level", "any listed findings", "required credentials, permissions, or environment variables", and "owner, source, version, changelog, downloads, stars, and other trust signals". Its advice is short: "Install only content you understand and trust."

Audit status tells you how to react to the result.

Risk level answers a different question. The page frames the two as a contrast: risk level asks "How much power is here?" while audit status asks "What should I do with this result?" Low means "Little sensitive authority or user impact was found", Medium that "The release has meaningful authority, such as account access or data changes", and High that "The release has high-impact authority, severe findings, or malicious signals."

Each listing links to its full audit:

/<owner>/skills/<slug>/security-audit

A Pass is not a clean bill of health. The same page warns that audits "are not a guarantee that a release is risk-free" and that "A Pass is reassuring, but it does not replace your own judgment."

#The twenty, at a glance

Downloads and stars below were read from each skill's ClawHub listing on 19 September 2026. They are registry counters, not ratings, and they move.

Slugs are not unique across publishers. The ClawHub HTTP API documentation says "Each item is identified by the owner-qualified pair ownerHandle/slug; slugs are not globally unique across publishers", which is why every command in this guide carries the owner handle.

#Communication Skills (1-4)

These skills let your agent work inside the messaging platforms and inboxes you already use.

#1. Discord

@steipete/discord sits in ClawHub's communication category. Its listing summary says to reach for it "when you need to control Discord from Clawdbot via the discord tool: send messages, react, post or upload stickers, upload emojis, run polls, manage threads/pins/search, fetch permissions or member/role/channel info, or handle moderation actions in Discord DMs or channels."

Listing: clawhub.ai/steipete/skills/discord

#2. WhatsApp Business

@byungkyu/whatsapp-business is filed under integrations and automation. The listing describes it as a "WhatsApp Business API integration with managed OAuth. Send messages, manage templates, and handle conversations." Its own summary also sets the default posture it expects from your agent: "Default to read and list calls, and confirm every write or new connection with the user."

Listing: clawhub.ai/byungkyu/skills/whatsapp-business

#3. AgentMail

@adboio/agentmail is in the communication category and gives an agent its own inbox rather than access to yours. The listing calls it an "API-first email platform designed for AI agents" that lets you "Create and manage dedicated email inboxes, send and receive emails programmatically, and handle email-based workflows with webhooks and real-time events."

Listing: clawhub.ai/adboio/skills/agentmail

#4. Telegram

OpenClaw already speaks Telegram for basic messaging. @codedao12/telegram, in the integrations category, is aimed at bot workflows instead: the listing describes an "OpenClaw skill for designing Telegram Bot API workflows and command-driven conversations using direct HTTPS requests (no SDKs)."

Listing: clawhub.ai/codedao12/skills/telegram

#Development Skills (5-8)

Four skills that sit where most engineering work already happens: the repo, the tracker, the pipeline and the framework docs.

#5. GitHub

@steipete/github had the second highest download count in this list on 19 September 2026, behind only Skill Vetter. It is an integrations skill that wraps the GitHub CLI rather than a bespoke API client: "Interact with GitHub using the gh CLI. Use gh issue, gh pr, gh run, and gh api for issues, PRs, CI runs, and advanced queries."

Listing: clawhub.ai/steipete/skills/github

#6. Linearis

@whoisnnamdi/linearis is an integrations skill built around a "Linear.app CLI for issue tracking." Its listing says to "Use for listing, creating, updating, and searching Linear issues, comments, documents, cycles, and projects", and that it is "Optimized for LLM agents with JSON output."

Listing: clawhub.ai/whoisnnamdi/skills/linearis

#7. CI/CD Pipeline

@gitgoodordietrying/cicd-pipeline carries three categories on ClawHub: development, automation and operations. The listing scopes it to GitHub Actions work: "Create, debug, and manage CI/CD pipelines with GitHub Actions", covering "workflow syntax, common patterns, secrets management, caching, matrix builds, and troubleshooting."

Listing: clawhub.ai/gitgoodordietrying/skills/cicd-pipeline

#8. Nextjs Expert

@jgarrison929/nextjs-expert is the framework reference of the group. Its listing says to invoke it "when building Next.js 14/15 applications with the App Router", for "routing, layouts, Server Components, Client Components, Server Actions, Route Handlers, authentication, middleware, data fetching, caching, revalidation, streaming, Suspense, loading states, error boundaries, dynamic routes, parallel routes, intercepting routes, or any Next.js architecture question." Check the version range against the Next.js release you actually run before you trust its answers.

Listing: clawhub.ai/jgarrison929/skills/nextjs-expert

#Research Skills (9-11)

Three ways to let the agent gather information rather than recall it.

#9. Playwright MCP

@spiceman161/playwright-mcp is filed under research and was the most downloaded of the three research entries on 19 September 2026. It provides "Browser automation via Playwright MCP server. Navigate websites, click elements, fill forms, extract data, take screenshots, and perform full browser automation workflows."

Listing: clawhub.ai/spiceman161/skills/playwright-mcp

#10. Deep Research

@bird-frank/deep-research-plan is a research skill for long-form investigation: "Automated deep research that performs comprehensive multi-source investigation and produces detailed reports with citations." It was nineteenth of the twenty by downloads on 19 September 2026, so read the SKILL.md closely before you lean on it.

Listing: clawhub.ai/bird-frank/skills/deep-research-plan

#11. Perplexity Search

@baofeng-tech/perplexity-search-aisa is a research skill covering "Perplexity Sonar search and answer generation through AIsa." The listing scopes it narrowly: use it "when the task is specifically to call Perplexity Sonar, Sonar Pro, Sonar Reasoning Pro" or the other Sonar models, rather than as a replacement for the agent's ordinary web search.

Listing: clawhub.ai/baofeng-tech/skills/perplexity-search-aisa

#Productivity Skills (12-15)

Notes, tasks, schedules and the jobs that run while you are asleep.

#12. Notion

@steipete/notion is the only productivity entry here, and was fourth by downloads of the twenty on 19 September 2026. Its summary is one line: "Notion API for creating and managing pages, databases, and blocks."

Listing: clawhub.ai/steipete/skills/notion

#13. Obsidian

@steipete/obsidian is filed under automation and works against a plain vault on disk: "Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli." It was third by stars in this list on 19 September 2026, behind Skill Vetter and GitHub.

Listing: clawhub.ai/steipete/skills/obsidian

#14. Cron and Scheduling

@gitgoodordietrying/cron-scheduling covers the automation and development categories, and it is the closest documented thing to a workflow builder in this list: "Schedule and manage recurring tasks with cron and systemd timers." Its listing names the failure case it is built for, "debugging why a scheduled task didn't run", alongside "timezone-aware scheduling, monitoring failed jobs" and "implementing retry patterns".

Pair it with the recipes in the Cron Job Recipes guide.

Listing: clawhub.ai/gitgoodordietrying/skills/cron-scheduling

#15. Google Calendar

@hith3sh/google-calendar-scheduling is an integrations skill that reads and writes a real calendar: "Check Google Calendar calendars, find free time, schedule meetings, and update events via the Google Calendar API."

Listing: clawhub.ai/hith3sh/skills/google-calendar-scheduling

#Voice and Media Skills (16-17)

Two skills that take the agent past text.

#16. Phone Voice Integration

@cortexuvula/phone-voice is an integrations skill that puts your agent on a phone line: "Connect ElevenLabs Agents to your OpenClaw via phone with Twilio. Includes caller ID auth, voice PIN security, call screening, memory injection, and cost tracking." It depends on two third-party accounts, so read their terms and pricing before wiring it up.

Listing: clawhub.ai/cortexuvula/skills/phone-voice

#17. OpenAI Whisper API

@steipete/openai-whisper-api is a creative skill with one job: "Transcribe audio via OpenAI Audio Transcriptions API (Whisper)." Point it at a recording, a voice note or a podcast episode you have already downloaded, and summarize the transcript with the agent afterwards.

Listing: clawhub.ai/steipete/skills/openai-whisper-api

#Security Skills (18-20)

Security skills protect your agent, your skills, and your infrastructure. All three below sit in ClawHub's security category.

#18. Security Audit Toolkit

@gitgoodordietrying/security-audit-toolkit points at your code rather than your agent: "Audit codebases and infrastructure for security issues." The listing covers "scanning dependencies for vulnerabilities, detecting hardcoded secrets, checking OWASP top 10 issues, verifying SSL/TLS, auditing file permissions, or reviewing code for injection and auth flaws."

Listing: clawhub.ai/gitgoodordietrying/skills/security-audit-toolkit

#19. ClawDefender

@nukewire/clawdefender points at the untrusted text your agent reads. It is a "Security scanner and input sanitizer for AI agents" that "Detects prompt injection, command injection, SSRF, credential exfiltration, and path traversal attacks", for cases including "processing external input like emails, calendar events, Trello cards, or API responses" and "validating URLs before fetching".

Listing: clawhub.ai/nukewire/skills/clawdefender

#20. Skill Vetter

@spclaudehome/skill-vetter points at the other skills. Its listing describes "Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub [sic], GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns."

Listing: clawhub.ai/spclaudehome/skills/skill-vetter

#Installing Multiple Skills at Once

There is no documented way to install several skills in one command. The skills CLI page lists every accepted form of install, and none of them takes a list of slugs, a bundle name or a --bundle flag. Installing five skills means running the command five times:

openclaw skills install @steipete/discord
openclaw skills install @byungkyu/whatsapp-business
openclaw skills install @adboio/agentmail
openclaw skills install @codedao12/telegram
openclaw skills install @steipete/github

By default each of those "installs into the active workspace skills/ directory". Add --global to target the shared managed skills directory instead, and --version <version> to pin a release.

#Checking Your Installed Skills

After installing, list what the Gateway actually loaded:

There is no --status flag. The documented ones are --eligible, --verbose, --json and --agent <id>, and list is the "Default action when no subcommand is provided", so bare openclaw skills prints the same table. No documented output column carries a star score or a badge, so a screenshot showing one is not describing this CLI.

If a skill you installed is missing from the list, ask why rather than reinstalling:

check "reports missing prerequisites independently of agent exclusion", so a skill lacking a binary, an environment variable or a config value shows up under missing requirements instead of vanishing silently.

#The awesome-openclaw-skills Repository

Beyond ClawHub itself, the community maintains a curated list on GitHub at VoltAgent/awesome-openclaw-skills, which carried 52,663 stars when we read it on 19 September 2026.

#What Makes It Special

It is an index, not a second registry. The contributing guide is explicit that every entry "must already be published" on ClawHub, that the repository curates "links only", and that each entry points at a skill's ClawHub page in the form https://clawhub.ai/author/skill-name. That makes it a browsing surface over the same listings you would install from anyway.

#How to Use It

gh repo clone VoltAgent/awesome-openclaw-skills

The list is split into a README.md plus a categories/ folder. These are some of the category files in that folder as of 19 September 2026:

awesome-openclaw-skills/
  README.md
  CONTRIBUTING.md
  categories/
    browser-and-automation.md
    calendar-and-scheduling.md
    communication.md
    devops-and-cloud.md
    git-and-github.md
    notes-and-pkm.md
    search-and-research.md
    security-and-passwords.md
    speech-and-transcription.md
    web-and-frontend-development.md

Note that these are the awesome list's own categories, which are not the same set as ClawHub's 14 browse category slugs. Understanding Skills has the ClawHub list.

#Contributing Your Own Skills

1. Publish your skill to ClawHub with the standalone clawhub CLI: npm i -g clawhub, then clawhub login, then clawhub skill publish ./my-skill (add --owner <handle> only when publishing under an org or user publisher handle you have access to)
2. Fork the awesome-openclaw-skills repo
3. Add your skill to the matching category in README.md
4. Open a pull request titled Add skill: author/skill-name with the ClawHub link in the description

The bar the maintainers publish is about provenance and adoption, not a rating. The contributing guide requires that "The skill's tests on ClawHub must be passing", that the skill "Has documentation (SKILL.md)", and that the "Skill must have real community usage", adding that "Brand new skills are not accepted" and that there are "No crypto, blockchain, DeFi, or finance-related skills for now". See Publishing to ClawHub for the publish side.

#Recommended Skill Combinations

These are stacks we would assemble from the twenty above, not packages you can install in one step.

#Next Steps

• Understanding Skills: how skills work under the hood
• Building Custom Skills: create your own skills
• Skill Security Audit: evaluate skills before installing
• Publishing to ClawHub: share your skills with the community
• Cost Optimization: manage costs across multiple skills