← Back to Articles
General7 min read

OpenClaw Tools List: A Full Guide to All Available Tools

ClawMakers Team·

OpenClaw Tools List: A Full Guide to All Available Tools

OpenClaw is a powerful AI agent gateway that connects your favorite messaging apps to AI assistants. At the heart of OpenClaw’s capabilities are its tools—pre-built functions that allow AI agents to interact with your system, perform actions, and automate tasks across platforms.

This guide provides a comprehensive overview of all available OpenClaw tools, organized by category. Whether you’re building workflows, automating messaging, or controlling your desktop, this reference will help you understand what each tool does and how to use it effectively.

What Are OpenClaw Tools?

OpenClaw tools are specialized functions that AI agents can call to perform actions outside the chat interface. Instead of relying on unstructured text commands, tools provide structured, reliable access to system capabilities—like running shell commands, controlling a browser, sending messages, or scheduling tasks.

Tools are designed to be safe, predictable, and auditable. You can control which tools are available through configuration, allowing you to create tailored experiences for different agents or use cases.

Core Interaction & Session Management

These tools handle communication and session control—essential for any agent that interacts with users or other services.

message

Purpose: Send messages across any connected chat platform.

Supported Channels: WhatsApp, Telegram, Discord, Slack, Signal, iMessage (BlueBubbles), Microsoft Teams, IRC, and more.

Capabilities:

  • Send text, images, videos, audio, documents
  • Reply to specific messages
  • Add reactions (emojis)
  • Create and vote in polls
  • Pin/unpin messages
  • Manage channels (create, edit, list)

Example Use Cases:

  • Responding to customer support queries
  • Broadcasting updates to a team
  • Automating social media posts

sessions_list, sessions_history, sessions_send, sessions_spawn

These tools allow agents to manage and communicate between sessions:

  • sessions_list: View active agent sessions
  • sessions_history: Retrieve conversation history
  • sessions_send: Send a message to another session
  • sessions_spawn: Launch a background sub-agent to handle complex tasks

Use Case: Delegate long-running tasks like data processing or research to a separate agent while keeping the main conversation flowing.

session_status

Check the current session’s model, usage statistics, and runtime environment. You can also temporarily override the model for the current session.

File & Workspace Operations

Manipulate files and directories within the OpenClaw workspace.

read

Read the contents of a file. Supports both text files and images (which are returned as attachments).

write

Create or overwrite a file. Automatically creates parent directories if they don’t exist.

edit

Make precise edits by replacing exact text matches. Ideal for updating configuration files or code without full content rewriting.

Tip: Combine with read and write for safe content transformation when exact match isn’t possible.

Command Execution & Process Control

Run shell commands and manage background processes—useful for automation and integration with existing CLI tools.

exec

Execute shell commands in the workspace.

Options:

  • yieldMs: Automatically background after a delay
  • background: Run immediately in background
  • timeout: Kill process if it runs too long
  • pty: Enable terminal mode (TTY) for interactive programs
  • host: Run on sandbox, gateway, or a remote node
  • elevated: Request higher privileges (requires approval)

Security Note: By default, exec runs in a sandbox. To run on the host, explicit configuration is required, often with approval workflows.

process

Manage background exec sessions:

  • poll: Check status and get output
  • log: Retrieve logs with offset/limit
  • send-keys: Simulate keyboard input (e.g., C-c for Ctrl+C)
  • paste: Inject text into stdin
  • kill: Terminate the process

Use Case: Monitor a long-running build or deployment and respond if errors occur.

Automation & Scheduling

Automate recurring tasks and manage the OpenClaw environment.

cron

Schedule and manage recurring tasks.

Actions:

  • add: Create a new scheduled job
  • list: View all jobs
  • update/remove: Modify or delete jobs
  • run: Trigger a job immediately
  • runs: Check execution history

Use Cases:

  • Daily system checks
  • Weekly report generation
  • Periodic monitoring of service status

gateway

Control the OpenClaw Gateway process:

  • restart: Reload the service
  • config.apply: Update configuration and restart
  • update.run: Apply updates and restart

Use Case: Automate configuration updates or deploy new agent versions.

Browser & UI Automation

Control a dedicated browser instance for web automation, testing, or data extraction.

browser

Full browser control with a dedicated, isolated Chromium profile.

Key Features:

  • Navigation: Open, close, and switch tabs
  • Interaction: Click, type, hover, drag, select
  • Snapshot: Extract accessibility tree or AI-friendly UI representation
  • Capture: Take screenshots (full page or element)
  • State: Manage cookies, localStorage, geolocation
  • Debugging: Console logs, network traces, error tracking

Automation Workflow:

  1. browser snapshot to get interactive element references
  2. browser act to click or type using those references
  3. browser screenshot to verify results

Security: The managed browser runs in isolation, separate from your personal browsing profile.

canvas

Control GUIs on paired devices (macOS, iOS) via the Canvas—a headless automation surface.

Use Case: Automate desktop applications that don’t have CLI interfaces.

nodes

Manage paired nodes (mobile or desktop):

  • Send notifications
  • Capture camera or screen
  • Run remote shell commands
  • Get device location

Use Case: Trigger actions on remote machines or mobile devices.

Web Research & Data

Quickly gather information from the web without launching a full browser.

web_search

Search the web using either:

  • Brave Search API (default, requires API key)
  • Perplexity Sonar (via OpenRouter or direct API)

Returns structured results with titles, URLs, and snippets.

web_fetch

Fetch a URL and extract readable content (HTML → markdown/text) using Readability.js.

Fallback: If JavaScript is required, it can use Firecrawl for dynamic content.

Use Case: Pull documentation, blog posts, or news articles into context.

Image & Multimedia

Work with visual and audio content.

image

Analyze images using a dedicated vision model (independent of the chat model).

Requires agents.defaults.imageModel to be configured.

Use Case: Describe screenshots, extract text from images, or analyze diagrams.

tts

Convert text to speech and return an audio file.

Use Case: Create voice responses for messaging apps that support audio.

Memory & Knowledge Retrieval

Access long-term memory and knowledge bases.

memory_search

Perform semantic search across:

  • MEMORY.md (curated long-term memory)
  • Daily logs (memory/YYYY-MM-DD.md)
  • Session transcripts

Best Practice: Always run memory_search before answering questions about prior decisions, preferences, or context.

memory_get

Retrieve specific lines from a memory file after discovery via memory_search.

Efficiency: Minimizes token usage by pulling only the relevant snippets.

Tool Configuration & Security

You can control tool availability through OpenClaw’s configuration system.

Tool Profiles

Set base allowlists using tools.profile:

  • minimal: Only session_status
  • coding: File, runtime, sessions, memory, image tools
  • messaging: Message and session tools
  • full: All tools enabled

Per-Provider Restrictions

Use tools.byProvider to restrict tools for specific LLM providers:

{
  "tools": {
    "byProvider": {
      "google-antigravity": { "profile": "minimal" }
    }
  }
}

This ensures that even powerful models can’t access sensitive tools unless explicitly allowed.

Conclusion

OpenClaw’s tool system transforms AI agents from chatbots into capable automation partners. From sending messages and running commands to controlling browsers and scheduling tasks, these tools give you fine-grained control over what your agents can do.

By understanding and leveraging these tools, you can build powerful, reliable workflows that save time and scale your operations—whether you’re managing a community, running a business, or automating your personal productivity.

ℹ️ Pro Tip: Combine tools like web_search + web_fetch + message to create a research agent that finds, summarizes, and shares information—all autonomously.

Check the official OpenClaw documentation for in-depth guides and API references.

This article was generated by Halie, Jorden’s AI assistant, using OpenClaw v2026.2.14.

Enjoyed this article?

Join the ClawMakers community to discuss this and more with fellow builders.

Join on Skool — It's Free →