OpenClaw turns your AI into an actual operator — not just a chatbot that generates text, but an agent that reads files, sends messages, controls browsers, deploys code, and manages your digital life. But out of the box, it's a foundation. Skills are what make it powerful.
This guide covers the skills worth installing right now, organized by what they actually do. We tested each one in real workflows — not just "does it install" but "does it save time."
🧑💻 Coding & Development
coding-agent
What it does: Delegates coding tasks to Claude Code, Codex, or Pi agents via background processes. Your agent can build features, review PRs, refactor codebases — all autonomously.
Why it matters: This is the skill that turns OpenClaw from "assistant" to "junior developer." Point it at a GitHub issue, and it will clone the repo, understand the codebase, write the fix, run tests, and open a PR. It handles the context management that makes coding agents actually work — feeding the right files to the right sub-agent at the right time.
Price: Free (built-in)
github
What it does: Full GitHub operations via the gh CLI — issues, PRs, CI runs, code review, API queries.
Why it matters: Combined with coding-agent, this creates a loop: find issue → write fix → open PR → monitor CI → respond to review comments. All hands-free. Also great for triage — "show me all open issues labeled bug, sorted by most recent."
Price: Free (built-in)
gh-issues
What it does: Fetches GitHub issues, spawns sub-agents to implement fixes, opens PRs, then monitors and addresses PR review comments. Supports labels, milestones, and assignee filters.
Why it matters: This is the "set and forget" version of GitHub automation. Point it at a repo with a bug backlog and let it work through issues sequentially. Each issue gets its own sub-agent with full context.
Price: Free (built-in)
🔍 Research & Web
deep-research-pro
What it does: Multi-source deep research agent. Searches the web, synthesizes findings, and delivers cited reports. No API keys required.
Why it matters: When you need actual research — not a summary of one page, but a synthesis across multiple sources with citations — this skill does it. It searches, reads, cross-references, and produces a report you can actually use for decision-making.
Price: Free
summarize
What it does: Summarizes or extracts text/transcripts from URLs, podcasts, and local files. Great fallback for YouTube/video transcription.
Why it matters: "Summarize this podcast" or "what does this 40-page PDF say" — both handled. It extracts the content first, then summarizes with the LLM. Works on YouTube videos, podcast RSS feeds, and local files.
Price: Free (built-in)
hn-extract
What it does: Extracts a Hacker News post (article + comments) into clean Markdown for quick reading or LLM input.
Why it matters: HN comments often contain more insight than the article itself. This skill pulls both the linked article AND the discussion thread into one clean document your agent can analyze.
Price: Free
📋 Productivity & Personal
apple-reminders
What it does: Manages Apple Reminders via the remindctl CLI — list, add, edit, complete, delete. Supports lists, date filters, and JSON output.
Why it matters: "Remind me to follow up with the client on Thursday" — done. Your agent writes directly to Apple Reminders, which syncs to your iPhone. No extra app needed.
Price: Free (built-in, macOS only)
apple-notes
What it does: Manages Apple Notes — create, view, edit, delete, search, move, and export notes.
Why it matters: Your agent can save research results, meeting notes, or ideas directly into Apple Notes where they sync across all your devices. Simple, native, reliable.
Price: Free (built-in, macOS only)
things-mac
What it does: Manages Things 3 via CLI — add/update projects and todos, read/search/list from the local Things database.
Why it matters: If you use Things for task management, this turns your agent into a planning assistant. "Break this project into tasks and add them to Things" — the agent creates the project with subtasks, deadlines, and tags.
Price: Free (built-in, requires Things 3)
obsidian
What it does: Works with Obsidian vaults (plain Markdown notes) and automates via obsidian-cli.
Why it matters: For knowledge workers with Obsidian vaults, this is transformative. Your agent can search your vault, create linked notes, update daily journals, and maintain a knowledge graph — all while respecting your existing folder structure.
Price: Free (built-in)
gog (Google Workspace)
What it does: Gmail, Calendar, Drive, Contacts, Sheets, and Docs — all via CLI.
Why it matters: This is the productivity swiss army knife. "Check my email for anything urgent," "what's on my calendar tomorrow," "create a spreadsheet with these numbers" — all through one skill. The calendar integration is especially useful for proactive agents that check your schedule during heartbeats.
Price: Free (built-in)
🤖 Automation & Scraping
peekaboo
What it does: Captures and automates macOS UI with the Peekaboo CLI — take screenshots, read screen content, interact with UI elements.
Why it matters: When there's no API and no CLI, there's the screen. Peekaboo lets your agent see and interact with any macOS application. It's the ultimate fallback for automation.
Price: Free (built-in, macOS only)
What it does: Browse, search, post, and moderate Reddit. Read-only works without auth; posting/moderation requires OAuth.
Why it matters: Market research, community engagement, content distribution — Reddit is where technical audiences hang out. Your agent can monitor subreddits, draft responses, and even post (with your approval).
Price: Free (built-in)
x-automation
What it does: Identifies trends on X (Twitter), generates engaging content, and posts it.
Why it matters: Social media presence on autopilot — but not the spammy kind. The skill identifies trending topics in your niche and generates opinionated content that sounds human. You review and approve before posting.
Price: Free (built-in)
🎤 Voice & Media
openai-whisper / openai-whisper-api
What it does: Local speech-to-text (no API key needed) or API-based transcription.
Why it matters: Two options: free local transcription (slower but private) or API transcription (fast, $0.006/min). Perfect for meeting recordings, voice memos, and podcast transcription.
Price: Free (local) / API costs apply
📈 SEO & Marketing
aeo-prompt-research-free
What it does: Discovers which AI prompts and topics matter for Answer Engine Optimization. Crawls websites, analyzes positioning, generates prioritized prompts.
Why it matters: As AI-generated answers replace traditional search results, optimizing for "what does ChatGPT say about your brand" becomes critical. This skill does that research automatically. No API keys required.
Price: Free
google-search-console
What it does: Google Search Console API integration with managed OAuth. Query search analytics, manage sitemaps, and monitor site performance.
Why it matters: "How are my rankings this week?" — answered instantly. Your agent can pull search analytics data and spot trends without you logging into GSC.
Price: Free
🔧 Infrastructure & DevOps
aws-infra
What it does: Chat-based AWS infrastructure assistance using AWS CLI — EC2, S3, IAM, Lambda, ECS/EKS, RDS, CloudWatch, billing, and more.
Why it matters: "What's running in my AWS account right now?" "How much did I spend this month?" "Spin up an EC2 instance for testing." All safe — the skill confirms before any write/destructive action.
Price: Free
1password
What it does: 1Password CLI integration — read secrets, inject into commands, manage vaults.
Why it matters: Secure secret management for agent workflows. Instead of hardcoding API keys, your agent can pull them from 1Password on demand. Essential for production setups.
Price: Free (requires 1Password subscription)
💰 Finance & Data
finance_lite
What it does: Daily macro + market brief with critical-headline triage, explicit source/freshness notes, and graceful fallback behavior.
Why it matters: Morning briefing: "What moved in markets today and why?" Gets you the essential context in 30 seconds instead of 30 minutes of scanning Bloomberg and CNBC.
Price: Free
Getting Started
Most built-in skills work immediately after installing OpenClaw. For community skills, install via ClawHub:
# Install a skill from ClawHub clawhub install skill-name # List installed skills clawhub list # Update all skills clawhub update
Browse the full catalog at TutuoAI Marketplace → or the machine-readable index at /api/tool-discovery.json.
This guide is updated regularly. Last update: March 9, 2026. Skills listed here were tested with OpenClaw on macOS. Some skills are platform-specific (noted where applicable). Prices may change — check the linked docs for current pricing.