Live
BREAKING: OpenAI GPT-5.5 Instant is now the default ChatGPT model — 52.5% fewer hallucinations on high-stakes topics; new memory sources feature shows users exactly what context shaped their responsesBREAKING: Grok falls to 5th place in AI chatbot rankings — Claude surges 1,205% year-over-year as xAI loses 80+ staff and faces App Store removal threat over explicit image controversyBREAKING: Five major publishers sue Meta over Llama — Hachette, Macmillan allege millions of books pirated for AI training; model can reproduce verbatim passagesBREAKING: Trump administration strikes AI safety deals with Google DeepMind, Microsoft, and xAI — pre-release model review extended to three more major labsBREAKING: Anthropic deploys 10 AI agent templates for Wall Street — Claude Opus 4.7 leads Finance Agent benchmark at 64.37%, now integrates with Excel, PowerPoint, Word, and OutlookBREAKING: Trump White House weighs executive order to vet AI models before release — Anthropic's Mythos model reportedly triggered the policy reversalBREAKING: IBM Think 2026 — watsonx Orchestrate repositioned as agentic control plane, IBM Bob reaches GA, Concert Secure Coder embeds security in developer workflowBREAKING: WEF report — 94% of cyber leaders say AI is defining force in cybersecurity; strategic AI adopters cut breach costs by $1.9M and shorten lifecycle by 80 daysBREAKING: 2026 cyberattack analysis — 17-year-old breached 7M records to buy Pokemon cards; time-to-exploit collapsed from 700 days in 2020 to just 44 days in 2025BREAKING: MIT Technology Review — AI is becoming the primary interface for democratic participation, but institutions were not designed for this worldBREAKING: Pentagon signs AI military deals with OpenAI, Google, Nvidia, Microsoft, Amazon, SpaceX, and Reflection AI — Anthropic refuses, sues over autonomous weapons concernsBREAKING: May Day 2026 — labor movement faces existential threat as Amazon plans to replace 500,000+ jobs with robots and AI automationBREAKING: AI research undergoes great pivot — focus shifts from model-centric breakthroughs to system-level deployment and autonomous scientific discoveryBREAKING: Northwestern study reveals AlphaFold2 expanded structural biology rather than replacing it — human-AI collaboration model offers template for futureBREAKING: PNNL scientists use machine learning to optimize nuclear waste vitrification at Hanford — could save hundreds of millions and reduce project timeline by yearsBREAKING: Meta raises AI capex to $125-145B, Google to $180-190B — trillion-dollar question: is the spending actually working?BREAKING: OpenAI launches GPT-5.5 agentic AI — AWS and Databricks announce managed agent services powered by GPT-5.5BREAKING: IBM launches Bob, end-to-end SDLC AI platform — $20/month Pro tier, multi-model orchestration, enterprise governanceBREAKING: Roblox Indonesia implements mandatory facial scanning for users under 16 — privacy advocates raise concernsBREAKING: DW investigation: AI industry burning trillions with no clear path to profitability — financial analysts warn of bubbleBREAKING: Microsoft drops exclusive OpenAI license — OpenAI now free to work with Amazon, Google, and any cloud providerBREAKING: China blocks Meta's $2B Manus acquisition — Beijing orders deal unwound in major cross-border AI tech rulingBREAKING: Musk vs. Altman trial begins — nine-person jury seated in Oakland, $134B OpenAI lawsuit opens TuesdayBREAKING: Big Tech's $600B AI earnings reckoning — Alphabet, Microsoft, Meta, Amazon all report WednesdayBREAKING: OpenAI proposes 4-day workweek, robot tax, and public AI wealth fund in sweeping economic policy blueprintBREAKING: Congress racing to reform FISA Section 702 before AI supercharges warrantless surveillance of AmericansBREAKING: Google reveals 75% of all new code is now AI-generated — up from just 25% eighteen months agoBREAKING: Google Gemini April Drop: native Mac app, AI music creation with Lyria 3 Pro, and Notebooks go liveBREAKING: Forbes AI 50 released — OpenAI leads at $182.6B funding, but vertical specialists are the real storyBREAKING: BCA Research warns AI trade entering 1999-style melt-up — S&P 500 could hit 9,200 before correctionBREAKING: OpenAI GPT-5.5 Instant is now the default ChatGPT model — 52.5% fewer hallucinations on high-stakes topics; new memory sources feature shows users exactly what context shaped their responsesBREAKING: Grok falls to 5th place in AI chatbot rankings — Claude surges 1,205% year-over-year as xAI loses 80+ staff and faces App Store removal threat over explicit image controversyBREAKING: Five major publishers sue Meta over Llama — Hachette, Macmillan allege millions of books pirated for AI training; model can reproduce verbatim passagesBREAKING: Trump administration strikes AI safety deals with Google DeepMind, Microsoft, and xAI — pre-release model review extended to three more major labsBREAKING: Anthropic deploys 10 AI agent templates for Wall Street — Claude Opus 4.7 leads Finance Agent benchmark at 64.37%, now integrates with Excel, PowerPoint, Word, and OutlookBREAKING: Trump White House weighs executive order to vet AI models before release — Anthropic's Mythos model reportedly triggered the policy reversalBREAKING: IBM Think 2026 — watsonx Orchestrate repositioned as agentic control plane, IBM Bob reaches GA, Concert Secure Coder embeds security in developer workflowBREAKING: WEF report — 94% of cyber leaders say AI is defining force in cybersecurity; strategic AI adopters cut breach costs by $1.9M and shorten lifecycle by 80 daysBREAKING: 2026 cyberattack analysis — 17-year-old breached 7M records to buy Pokemon cards; time-to-exploit collapsed from 700 days in 2020 to just 44 days in 2025BREAKING: MIT Technology Review — AI is becoming the primary interface for democratic participation, but institutions were not designed for this worldBREAKING: Pentagon signs AI military deals with OpenAI, Google, Nvidia, Microsoft, Amazon, SpaceX, and Reflection AI — Anthropic refuses, sues over autonomous weapons concernsBREAKING: May Day 2026 — labor movement faces existential threat as Amazon plans to replace 500,000+ jobs with robots and AI automationBREAKING: AI research undergoes great pivot — focus shifts from model-centric breakthroughs to system-level deployment and autonomous scientific discoveryBREAKING: Northwestern study reveals AlphaFold2 expanded structural biology rather than replacing it — human-AI collaboration model offers template for futureBREAKING: PNNL scientists use machine learning to optimize nuclear waste vitrification at Hanford — could save hundreds of millions and reduce project timeline by yearsBREAKING: Meta raises AI capex to $125-145B, Google to $180-190B — trillion-dollar question: is the spending actually working?BREAKING: OpenAI launches GPT-5.5 agentic AI — AWS and Databricks announce managed agent services powered by GPT-5.5BREAKING: IBM launches Bob, end-to-end SDLC AI platform — $20/month Pro tier, multi-model orchestration, enterprise governanceBREAKING: Roblox Indonesia implements mandatory facial scanning for users under 16 — privacy advocates raise concernsBREAKING: DW investigation: AI industry burning trillions with no clear path to profitability — financial analysts warn of bubbleBREAKING: Microsoft drops exclusive OpenAI license — OpenAI now free to work with Amazon, Google, and any cloud providerBREAKING: China blocks Meta's $2B Manus acquisition — Beijing orders deal unwound in major cross-border AI tech rulingBREAKING: Musk vs. Altman trial begins — nine-person jury seated in Oakland, $134B OpenAI lawsuit opens TuesdayBREAKING: Big Tech's $600B AI earnings reckoning — Alphabet, Microsoft, Meta, Amazon all report WednesdayBREAKING: OpenAI proposes 4-day workweek, robot tax, and public AI wealth fund in sweeping economic policy blueprintBREAKING: Congress racing to reform FISA Section 702 before AI supercharges warrantless surveillance of AmericansBREAKING: Google reveals 75% of all new code is now AI-generated — up from just 25% eighteen months agoBREAKING: Google Gemini April Drop: native Mac app, AI music creation with Lyria 3 Pro, and Notebooks go liveBREAKING: Forbes AI 50 released — OpenAI leads at $182.6B funding, but vertical specialists are the real storyBREAKING: BCA Research warns AI trade entering 1999-style melt-up — S&P 500 could hit 9,200 before correctionBREAKING: OpenAI GPT-5.5 Instant is now the default ChatGPT model — 52.5% fewer hallucinations on high-stakes topics; new memory sources feature shows users exactly what context shaped their responsesBREAKING: Grok falls to 5th place in AI chatbot rankings — Claude surges 1,205% year-over-year as xAI loses 80+ staff and faces App Store removal threat over explicit image controversyBREAKING: Five major publishers sue Meta over Llama — Hachette, Macmillan allege millions of books pirated for AI training; model can reproduce verbatim passagesBREAKING: Trump administration strikes AI safety deals with Google DeepMind, Microsoft, and xAI — pre-release model review extended to three more major labsBREAKING: Anthropic deploys 10 AI agent templates for Wall Street — Claude Opus 4.7 leads Finance Agent benchmark at 64.37%, now integrates with Excel, PowerPoint, Word, and OutlookBREAKING: Trump White House weighs executive order to vet AI models before release — Anthropic's Mythos model reportedly triggered the policy reversalBREAKING: IBM Think 2026 — watsonx Orchestrate repositioned as agentic control plane, IBM Bob reaches GA, Concert Secure Coder embeds security in developer workflowBREAKING: WEF report — 94% of cyber leaders say AI is defining force in cybersecurity; strategic AI adopters cut breach costs by $1.9M and shorten lifecycle by 80 daysBREAKING: 2026 cyberattack analysis — 17-year-old breached 7M records to buy Pokemon cards; time-to-exploit collapsed from 700 days in 2020 to just 44 days in 2025BREAKING: MIT Technology Review — AI is becoming the primary interface for democratic participation, but institutions were not designed for this worldBREAKING: Pentagon signs AI military deals with OpenAI, Google, Nvidia, Microsoft, Amazon, SpaceX, and Reflection AI — Anthropic refuses, sues over autonomous weapons concernsBREAKING: May Day 2026 — labor movement faces existential threat as Amazon plans to replace 500,000+ jobs with robots and AI automationBREAKING: AI research undergoes great pivot — focus shifts from model-centric breakthroughs to system-level deployment and autonomous scientific discoveryBREAKING: Northwestern study reveals AlphaFold2 expanded structural biology rather than replacing it — human-AI collaboration model offers template for futureBREAKING: PNNL scientists use machine learning to optimize nuclear waste vitrification at Hanford — could save hundreds of millions and reduce project timeline by yearsBREAKING: Meta raises AI capex to $125-145B, Google to $180-190B — trillion-dollar question: is the spending actually working?BREAKING: OpenAI launches GPT-5.5 agentic AI — AWS and Databricks announce managed agent services powered by GPT-5.5BREAKING: IBM launches Bob, end-to-end SDLC AI platform — $20/month Pro tier, multi-model orchestration, enterprise governanceBREAKING: Roblox Indonesia implements mandatory facial scanning for users under 16 — privacy advocates raise concernsBREAKING: DW investigation: AI industry burning trillions with no clear path to profitability — financial analysts warn of bubbleBREAKING: Microsoft drops exclusive OpenAI license — OpenAI now free to work with Amazon, Google, and any cloud providerBREAKING: China blocks Meta's $2B Manus acquisition — Beijing orders deal unwound in major cross-border AI tech rulingBREAKING: Musk vs. Altman trial begins — nine-person jury seated in Oakland, $134B OpenAI lawsuit opens TuesdayBREAKING: Big Tech's $600B AI earnings reckoning — Alphabet, Microsoft, Meta, Amazon all report WednesdayBREAKING: OpenAI proposes 4-day workweek, robot tax, and public AI wealth fund in sweeping economic policy blueprintBREAKING: Congress racing to reform FISA Section 702 before AI supercharges warrantless surveillance of AmericansBREAKING: Google reveals 75% of all new code is now AI-generated — up from just 25% eighteen months agoBREAKING: Google Gemini April Drop: native Mac app, AI music creation with Lyria 3 Pro, and Notebooks go liveBREAKING: Forbes AI 50 released — OpenAI leads at $182.6B funding, but vertical specialists are the real storyBREAKING: BCA Research warns AI trade entering 1999-style melt-up — S&P 500 could hit 9,200 before correction
GuidesSetup
advanced9 min read

OpenClaw Skills Setup Guide

---

S
By SUPERBASH_

OpenClaw Skills Setup Guide

Overview

Skills are the most important feature you should set up on your OpenClaw agent. They allow your bot to perform tasks exactly to your specifications, consistently and reliably. This guide teaches you how to create, manage, and optimize skills for your workflow.

Video Reference: Watch on YouTube
Duration: ~7 minutes
Difficulty: Beginner


What Are Skills?

Skills are reusable instruction sets that tell your agent exactly how to perform specific tasks. Think of them as:

  • Recipes for your agent to follow
  • Muscle memory that persists across sessions
  • Custom commands tailored to your needs

Why Skills Matter

Without Skills: You explain your preferences every session

With Skills: Your agent remembers and applies your preferences automatically


The Core Problem Skills Solve

Your Agent "Dies" Every Day

This is the best way to understand it:

  • Your agent's context window resets regularly
  • It "forgets" everything from previous sessions
  • It has to relearn your preferences each time

Skills = Persistent Memory

Skills give your agent knowledge that survives resets:

  • How you like presentations formatted
  • Your research methodology
  • Your writing style
  • Your workflow preferences

Creating Your First Skill

Method 1: Conversational Creation (Recommended)

You don't need to write skills yourself—just talk to your agent.

Step 1: Describe What You Want

I need you to create daily presentations about AI and crypto news. 
Here's how I want them:
- Research latest news from the past 24 hours
- Create 5-7 slides
- Use light theme with blue accents
- Include sources at the end
- Save to ~/presentations/daily/

Step 2: Let Your Agent Work

Your agent will:

  1. Understand your requirements
  2. Perform the task
  3. Ask clarifying questions if needed

Step 3: Refine Through Iteration

After seeing the result:

Good, but make the slides more concise. 
Each slide should have max 3 bullet points.
Update your skill with this preference.

Step 4: Save as Skill

Save this as a skill called "daily_presentation" 
so you can do this automatically every day.

[Image blocked: Skill Creation Confirmed]

Method 2: Explicit Skill Request

Direct Approach:

Create a skill for making presentations with these specs:
- Title slide with date
- 5 content slides max
- Light theme
- Sources slide at end
- Save to ~/presentations/

Your agent will create the skill immediately.


Real-World Example: Presentation Skill

The Workflow

Morning Routine:

  1. Agent runs at 8:00 AM (via cron job)
  2. Invokes "daily_presentation" skill
  3. Researches latest news
  4. Creates presentation
  5. Saves to specified folder
  6. Notifies you when complete

No Manual Work Required: Everything happens automatically to your exact specifications.

The Skill Structure

Your agent creates something like this internally:

Skill: daily_presentation

Purpose: Create daily news presentation

Steps:
1. Research AI and crypto news from past 24 hours
2. Identify top 5 stories
3. Create presentation with:
   - Title slide (date)
   - 5 content slides (3 bullets each)
   - Light theme, blue accents
   - Sources slide
4. Save to ~/presentations/daily/YYYY-MM-DD.pptx
5. Notify user when complete

Dependencies: 
- deep_research skill
- presentation_template

You never see this—your agent manages it internally.


Viewing Your Skills

Ask Your Agent

Show me all my skills

or

Send me the presentation skill details

Your agent will display the skill specifications.

Understanding Skill Details

When you view a skill, you'll see:

  • Purpose: What the skill does
  • Steps: How it accomplishes the task
  • Preferences: Your specific requirements
  • Dependencies: Other skills it uses

[Image blocked: Skill Details Display]


Updating Skills

When to Update

  • You discover a better way to do something
  • Your preferences change
  • You notice consistent issues
  • You want to add new features

How to Update

Simple Updates:

Update your presentation skill: 
I prefer dark theme now, not light theme.

Detailed Updates:

Update the daily_presentation skill with these changes:
1. Increase to 7 slides instead of 5
2. Add a "Key Takeaways" slide before sources
3. Use dark theme with green accents
4. Include tweet-length summaries for each story

Your agent will modify the skill and confirm changes.


Skill Categories

Research Skills

Example: Deep research on specific topics

Create a skill for deep research that:
- Searches multiple sources
- Verifies information across sources
- Summarizes findings
- Includes citations
- Saves to ~/research/

Content Creation Skills

Example: Blog post writing

Create a skill for blog posts that:
- Researches topic thoroughly
- Writes in my style (conversational, technical)
- Includes code examples when relevant
- Adds SEO-friendly title and meta description
- Saves as markdown

Automation Skills

Example: Daily briefing

Create a skill for daily briefings that:
- Checks email for urgent items
- Reviews calendar for today
- Summarizes news in my industry
- Compiles into one document
- Sends to my Telegram

Analysis Skills

Example: Data analysis

Create a skill for analyzing my trading data that:
- Reads from Notion database
- Identifies patterns in winning trades
- Calculates key metrics
- Generates visual charts
- Saves report to ~/analysis/

Skills vs Memory

When to Use Skills

Repetitive tasks with consistent requirements
Specific workflows you use regularly
Quality standards that must be maintained
Complex processes with multiple steps

When to Use Memory

Personal facts about you
Project context that changes
Temporary preferences
Conversation history

Example Comparison

Skill: "How to create presentations"

  • Format specifications
  • Design preferences
  • File naming conventions
  • Save locations

Memory: "User is working on Q2 report"

  • Current project details
  • Temporary focus areas
  • Recent conversations

Advanced: Skill Dependencies

Skills can use other skills.

Example: Presentation Skill Uses Research Skill

Main Skill: daily_presentation

  • Calls: deep_research skill
  • Calls: format_slides skill
  • Calls: save_file skill

Benefit: Modular, reusable components

Creating Dependent Skills

Create a research skill that other skills can use:
- Takes a topic as input
- Searches multiple sources
- Verifies information
- Returns structured summary

Then:

Update my presentation skill to use the research skill 
for gathering information.

ClawHub: Skill Marketplace

What is ClawHub?

A marketplace where users share skills.

Should You Use It?

Pros:

  • Browse for inspiration
  • See what's possible
  • Quick start for common tasks

Cons:

  • Security concerns (past virus issues)
  • Generic skills may not fit your needs
  • Less control over implementation

Recommended Approach

  1. Browse for ideas (don't download)
  2. Understand what the skill does
  3. Ask your agent to create similar (customized for you)

Example:

I saw a skill on ClawHub for email management. 
Create a similar skill but customized for my workflow:
[describe your specific needs]

Troubleshooting

Skill Not Working

Problem: Skill fails or produces wrong results

Solutions:

  1. Ask agent to explain what went wrong
  2. Review skill specifications
  3. Update with corrections
  4. Test again
The presentation skill failed. 
Show me the error and fix it.

Skill Forgotten After Restart

Problem: Agent doesn't remember skill after restart

Solutions:

  1. Verify skill was saved (not just executed once)
  2. Ask agent to list all skills
  3. Recreate if necessary
List all my saved skills. 
If presentation skill is missing, recreate it.

Skill Produces Inconsistent Results

Problem: Sometimes works, sometimes doesn't

Solutions:

  1. Make specifications more explicit
  2. Add error handling to skill
  3. Include verification steps
Update the skill to verify each step completes 
before moving to the next.

Best Practices

1. Start Simple

Don't create complex skills immediately. Start with:

Create a skill to summarize articles I send you.

Then gradually add complexity:

Update the summary skill to also:
- Extract key quotes
- Identify main arguments
- Suggest related topics

2. Be Specific

Vague:

Make presentations look good

Specific:

Use dark theme, 24pt titles, 18pt body text, 
blue (#0066CC) for headings, white text on dark background

3. Test Thoroughly

Before relying on a skill:

Run the presentation skill 3 times with different topics 
to ensure consistency.

4. Document Your Skills

Keep a list:

Create a document listing all my skills, 
what they do, and when to use them.

5. Regular Maintenance

Monthly review:

Review all my skills. 
Which ones haven't been used in 30 days? 
Should we update or remove them?

Skill Ideas to Get Started

For Content Creators

  1. Video Script Writer: Research topic, write script, format for teleprompter
  2. Thumbnail Analyzer: Analyze successful thumbnails, suggest improvements
  3. SEO Optimizer: Optimize titles, descriptions, tags

For Developers

  1. Code Reviewer: Review code for bugs, suggest improvements
  2. Documentation Generator: Create docs from code comments
  3. Test Writer: Generate unit tests for functions

For Business

  1. Meeting Summarizer: Summarize meeting notes, extract action items
  2. Email Drafter: Draft professional emails based on bullet points
  3. Report Generator: Compile data into formatted reports

For Researchers

  1. Literature Review: Search papers, summarize findings, identify gaps
  2. Citation Manager: Format citations, check for duplicates
  3. Data Analyzer: Analyze datasets, generate visualizations

Next Steps

After creating your first skill:

  1. Use It Regularly: Skills improve with feedback
  2. Create More Skills: Build a library for your workflow
  3. Combine Skills: Create complex workflows from simple skills
  4. Share Knowledge: Help others by describing your successful skills

Related Resources

  • OpenClaw API Integration Guide [blocked]
  • OpenClaw Cron Jobs Setup [blocked]
  • Hermes Agent Curator Guide [blocked]

Final Thoughts

Skills transform your agent from a helpful assistant into a personalized automation system. The key is to start simple, iterate based on results, and gradually build a library of skills that match your exact workflow.

Remember: You don't need to write code. Just describe what you want in plain language, and your agent will create the skill for you.


Last Updated: 2026-05-06