🦞 OpenClaw Bootcamp
DAY 10 / 16
🔧
OpenClaw Bootcamp · Day 10

Tools & Skills:
Extend Your Agent

Your agent talks, remembers, delegates, and follows rules. But it can only do what its tools allow. Tools grant raw capabilities. Skills teach your agent how to use them. Today you learn both — and how to add more.

~20 Built-in Tools 53+ Bundled Skills 3000+ on ClawHub
🦞 OpenClaw Bootcamp
DAY 10 / 16
Where We Are & Where We’re Going

Day 10 Goals

Day 9 Checkpoint
  • Custom AGENTS.md with safety defaults and operating rules
  • Spawned your first sub-agent and watched delegation in action
  • Understand sandbox, tool policy, and elevated — the three security layers
01
Know the difference between Tools and Skills — and why it matters
02
Master the SKILL.md format, loading order, and per-agent configuration
03
Install from ClawHub, build a custom skill, and audit third-party code for security
🦞 OpenClaw Bootcamp
DAY 10 / 16
The Core Distinction

Tools vs Skills

Tools = Raw Capabilities

What your agent can do. Each tool is a function the runtime can execute: exec, read, write, edit, web_search, web_fetch, browser, message, cron, and the sessions family. ~20 built-in tools ship with every installation. Controlled by tools.allow in config.

📚

Skills = Instruction Manuals

How your agent uses tools. A Skill is a SKILL.md file with YAML frontmatter + markdown instructions that teach the agent when and how to apply its tools. Skills don’t add permissions — they’re guidance, not grants. 53 bundled skills, 3000+ on ClawHub.

~20
Built-in Tools
53+
Bundled Skills
3000+
ClawHub Skills
Key Rule

Skills never add permissions. If a tool is denied by tools.allow, no Skill can override that. For a Skill to work, three things must be true: the underlying tool is enabled (Configuration), any required bridge or CLI is installed (Installation), and the agent is authorized for the service (Authorization).

🦞 OpenClaw Bootcamp
DAY 10 / 16
What Ships Out of the Box

Core Built-in Tools

💻

exec

Shell execution — run terminal commands, get stdout/stderr. Most powerful and most dangerous. Requires careful allow-listing in production.

📁

read / write / edit

Filesystem operations — read, write, append, edit files. Foundation for memory writes, config changes, and all file-based workflows.

🔍

web_search / web_fetch

Search engines and HTTP requests. Gives your agent access to current information and any REST API endpoint.

🌐

browser

Browser automation — navigate pages, click elements, fill forms, extract content. Playwright-based headless interaction.

💬

message

Send messages to channels — Telegram, Discord, Slack, WhatsApp. The bridge between your agent and external platforms.

cron & sessions

Scheduling and session management — cron for timed tasks, sessions_list/history/send/spawn for managing agent runs and sub-agents.

🦞 OpenClaw Bootcamp
DAY 10 / 16
Anatomy of a Skill

SKILL.md Format & Loading

YAML Frontmatter
--- name: my-custom-skill description: One-line summary user-invocable: true # show as /command metadata: os: [linux, darwin] requiredBinaries: [node] env: [API_KEY] --- ## Instructions Use this when the user asks... Do NOT use for...
Required Fields

name and description are required. Everything else is optional. Frontmatter keys must be single-line — the parser doesn’t support multi-line values.

Loading Order (Precedence)
  • 01Workspace skills/
  • 02Workspace .agents/skills/
  • 03~/.agents/skills/
  • 04~/.openclaw/skills/
  • 05Bundled skills (53+)
  • 06skills.load.extraDirs
Snapshot Behavior

Skills are snapshotted when a session starts. If you edit a SKILL.md, you need to run /new or restart the gateway for changes to take effect.

🦞 OpenClaw Bootcamp
DAY 10 / 16
openclaw.json

Skill Configuration

skills.entries (Per-Skill)
"skills": { "entries": { "my-skill": { "enabled": true, "apiKey": { "source": "env", "id": "MY_API_KEY" }, "env": { "REGION": "us-east-1" } } } }
SecretRef

API keys use SecretRef objects ({ source: "env", id: "VAR_NAME" }) so secrets stay in environment variables, never in config files.

Loading & Install Options
  • skills.load.watch — Watch for file changes (takes effect on /new)
  • skills.load.extraDirs — Additional skill directories
  • skills.install.nodeManager — npm, pnpm, yarn, or bun
Per-Agent Skill Access
  • agents.defaults.skills — Inherited baseline for all agents
  • agents.list[].skills — Per-agent override

Explicit skill lists replace inherited defaults — they do NOT merge. If you set a skills array on a specific agent, it completely replaces the defaults.

🦞 OpenClaw Bootcamp
DAY 10 / 16
The Marketplace

ClawHub: 3000+ Skills

CLI Commands
clawhub search weather clawhub install skill-name clawhub update --all clawhub sync clawhub login
After Install

Skills become available after a /new session or gateway restart. Configure credentials in openclaw.json under skills.entries.

Each Listing Shows
  • Publisher identity & verification status
  • Install counts & community ratings
  • Open-source code availability
  • Compatibility & config requirements
Categories
Productivity Communication Development Finance Smart Home Browser
🦞 OpenClaw Bootcamp
DAY 10 / 16
The Attack Surface

Skill Security

The Docs Warn

“Skills are the most significant attack surface in any OpenClaw deployment.” A malicious Skill runs with the same permissions as the OpenClaw process — access to files, shell, API keys, network.

Known Attack Types
  • Credential theft & SSH directory exfiltration
  • Crypto wallet address substitution
  • Persistent remote access backdoors
  • Telemetry reporting to external servers
The Reality

ClawHub runs a built-in dangerous-code scanner that blocks critical findings, but malicious code can still slip through — npm lifecycle scripts execute during install. Skills auto-hide after 3+ unique user reports.

Your Defense
  • 01Read the source — most skills are 100–300 lines
  • 02Run in Docker with network restrictions
  • 03Verify the publisher — check GitHub reputation
  • 04Never install from unverified URLs
🦞 OpenClaw Bootcamp
DAY 10 / 16
Before Day 11

Day 10 Homework

  • 01

    List Your Installed Skills

    Run clawhub search to browse available skills. Then run /context list in chat to see how much skill definitions contribute to your context window. Remove anything you’re not using.

  • 02

    Install One Skill from ClawHub

    Run clawhub search to find something relevant to your workflow. Before installing, read the source code — remember, this is a security decision. Test it in a conversation.

  • 03

    Build a Custom SKILL.md

    Create a directory under ~/.openclaw/skills/ with a SKILL.md file. It can be simple — a unit converter, a quote generator, a task formatter. Focus on getting the “Use this when” instructions precise enough that the agent calls it at the right time.

  • 04

    Security Audit One Skill

    Pick any installed skill — built-in or third-party — and read its source code end to end. Check for hardcoded credentials, unexpected outbound connections, and unsanitized inputs. Build the audit habit now.

🦞 OpenClaw Bootcamp
DAY 10 / 16
🦞
Coming Up

Day 11: Integrations
Connecting to Real-World Services

Your agent has tools and skills. Now it needs to plug into the services you actually use every day — webhooks, authentication patterns, API connectors, and making your agent a first-class citizen in your existing toolchain. Day 11 wires it all together.

Webhooks Auth Patterns API Connectors