- 23
- February
In early 2026, the AI Agent trend is hotter than ever before. AI is no longer just a chatbot that answers questions — it has become an "intelligent agent" that can actually perform tasks on your behalf, whether reading emails, managing files, browsing the web, or even running commands on your own computer. And one of the most talked-about names is OpenClaw — an open-source AI Agent with more than 219,000 GitHub stars, making it the fastest-growing open-source project in history.
What Is OpenClaw?
OpenClaw is an open-source AI Agent (MIT License) that runs automated tasks on your own machine. It was created by Peter Steinberger, an Austrian developer who founded PSPDFKit, a PDF SDK company that generated approximately $100 million in revenue before being sold.
What sets OpenClaw apart from ordinary chatbots like ChatGPT or Claude is that OpenClaw has both "eyes" and "hands":
- Eyes — It can browse the web, read local files, read emails, and view calendars
- Hands — It can write files, send emails, run Shell commands, and interact with APIs
Most importantly, you can use it through familiar chat platforms right away — whether WhatsApp, Telegram, Discord, Slack, Signal, iMessage, Microsoft Teams, Matrix, or even Zalo. Simply send a message with your instructions and OpenClaw will carry them out automatically.
In short: OpenClaw = an AI that does not just "talk" but actually "does" — running on your own machine, controllable via Chat.
History — From Clawdbot to OpenClaw
OpenClaw's journey has been full of trademark disputes and multiple name changes. Here is the key timeline:
| Date | Event |
|---|---|
| Nov 2025 | Peter Steinberger created "Clawdbot" in roughly one hour by connecting a chat app to Claude Code, turning it into a personal AI Agent. |
| Jan 2026 | The project exploded in popularity thanks to the Moltbook trend (a social media forum for AI agents), with tens of thousands of downloads and forks. |
| 27 Jan 2026 | Anthropic raised a trademark complaint about "Clawd" being too similar to "Claude," forcing a rename to "Moltbot." |
| 30 Jan 2026 | Officially renamed to "OpenClaw" (permanent name) with a full project rebrand. |
| 14 Feb 2026 | Steinberger announced he was joining OpenAI and that OpenClaw would be transferred to an open-source foundation for the community to maintain. |
Key Features of OpenClaw
OpenClaw is not just an ordinary chatbot — it is a full-fledged AI Agent with a wide range of capabilities:
| Feature | Description |
|---|---|
| Read/Write Files | Read, edit, create, and delete files on your local machine, including folder management. |
| Run Shell Commands | Execute Terminal/Shell commands on the machine, such as installing packages, moving files, and running scripts. |
| Web Browsing | Open Chrome/Chromium to search for information, fill out forms, download files, and take screenshots of web pages. |
| Email Management | Read, compose, and send emails; triage inbox by priority; summarize long emails. |
| Calendar Management | View schedules, create new appointments, send reminders, and organize meetings. |
| Cron Jobs / Automation | Schedule recurring automated tasks, such as summarizing news every morning or backing up files every night. |
| Voice Integration | Supports wake word for voice commands (requires additional configuration). |
| Skills / Plugins | The ClawHub marketplace offers more than 3,286 skills, from sending LINE messages and controlling smart home devices to trading bots. |
| Multi-Agent Routing | Create multiple agents that work together — for example, one agent reads emails, passes them to a summarizer agent, then to a reply agent. |
| Persistent Memory 24/7 | Remembers context and data from previous conversations, enabling continuous operation without forgetting. |
Which LLMs Are Supported?
OpenClaw uses the Bring Your Own Model (BYOM) principle — you choose which LLM to use with your own API key, without being locked into any single provider:
| Provider | Supported Models | Notes |
|---|---|---|
| Anthropic | Claude Opus 4.6, Claude Sonnet 4 | Recommended for agent tasks (best Tool Use support) |
| OpenAI | GPT-5.2, GPT-4o, o3 | Popular choice with various API pricing tiers |
| DeepSeek | DeepSeek-V3, DeepSeek-R1 | Affordable, but data is sent to servers in China |
| Local Models | Ollama, vLLM, LM Studio | 100% Free — data never leaves your machine |
Key advantage: If you use a local model (e.g., Llama 3 via Ollama), your LLM cost is $0 because everything runs on your own machine with no API fees.
How to Install OpenClaw
Installing OpenClaw is relatively straightforward. It natively supports macOS and Linux, while Windows users can run it through WSL2:
Minimum Requirements
- Node.js 22+ (install via nvm or brew)
- macOS / Linux — natively supported
- Windows — use via WSL2 (Windows Subsystem for Linux)
- RAM — at least 4 GB (16 GB+ if using a local model)
Install with One-liner
curl -fsSL https://openclaw.ai/install.sh | bash
Install via npm
npm install -g openclaw@latest
Basic Configuration
After installation, set up the minimum configuration at ~/.openclaw/openclaw.json by specifying the LLM model you want to use:
{
"model": "claude-opus-4-6",
"provider": "anthropic",
"api_key": "sk-ant-xxxxx",
"chat_platform": "telegram"
}
Pricing
The OpenClaw software itself is 100% free and open-source, but actual costs depend on the LLM you choose:
| Item | Price | Notes |
|---|---|---|
| OpenClaw (Self-hosted) | Free (MIT License) | Self-install, self-maintain |
| LLM API Cost | Varies by provider | Claude ~$15-75/MTok, GPT-4o ~$2.5-10/MTok |
| Local Model (Ollama) | $0 | Completely free, but requires a powerful enough GPU |
| OpenClaw Cloud (Managed) | Starting at $39/month | No self-installation required, managed by a dedicated team (still in Early Access) |
Security Considerations (Very Important!)
While OpenClaw is a powerful tool, that power comes with serious security risks that every user must be aware of:
Security Warnings
1. CVE-2026-25253 (CVSS 8.8 — High Severity)
A Remote Code Execution (RCE) vulnerability was discovered that allows attackers to run code on your machine remotely simply by sending a message through a chat platform connected to OpenClaw.
2. 40,000+ Instances Exposed Without Authentication
Security researchers found that more than 40,000 OpenClaw instances have their Web UI exposed to the internet without any authentication, allowing anyone to take control of those AI Agents.
3. API Keys Stored in Plaintext
OpenClaw stores LLM provider API keys in a plaintext JSON file without encryption. If the machine is compromised, the API keys can be stolen immediately.
4. Malicious Code in ClawHub Skills
Investigations revealed that 341 skills on the ClawHub marketplace contain malicious code, ranging from API key theft to cryptocurrency miner installation.
5. Prompt Injection
An industry-wide problem that remains unsolved — malicious actors can embed hidden commands in emails, websites, or documents. When OpenClaw reads these, it may automatically execute those commands, such as exfiltrating confidential files or deleting critical data.
Multiple leading cybersecurity companies have issued warnings about this, including CrowdStrike, Microsoft, Cisco Talos, Kaspersky, and Trend Micro, all stating that AI Agents with Shell-level system access represent a high-level risk for organizations.
Recommendation: If using OpenClaw in an organization, it must be run inside a separate VM (Virtual Machine) that is not connected to the internal network. Never run it on a production workstation. Restrict file and Shell access to the absolute minimum.
Use Cases for Organizations
Despite the security concerns, when used carefully within an isolated environment, OpenClaw has high potential across many use cases:
1. Email Triage
Instead of spending 2+ hours reading and prioritizing emails every morning, OpenClaw can complete email triage in under 25 minutes by categorizing, summarizing content, and flagging emails that require urgent responses.
2. Customer Support (First-line)
Connect OpenClaw to your organization's WhatsApp or Telegram to answer basic customer questions such as order status, usage instructions, or business hours, then escalate to human staff when complex issues arise.
3. Client Onboarding
The client onboarding process that previously took 3-4 hours to prepare documents, create accounts, and send welcome emails can be reduced to 15 minutes by letting OpenClaw handle all the repetitive tasks.
4. IT Automation
Use OpenClaw for monitoring, log analysis, and server health checks by setting up cron jobs to run checks every hour, sending alerts via chat when issues are found, along with recommended fixes.
5. File Management & Reports
Instruct OpenClaw to gather data from multiple files, generate summary reports, convert document formats, or organize folders according to defined rules — completing tasks that used to take half a day in just minutes.
Saeree ERP and OpenClaw
Important note: Currently, Saeree ERP does not have built-in AI features, but this is in the near-term development roadmap. In the meantime, organizations using Saeree ERP can use OpenClaw as an external supplementary tool — for example, to help summarize PDF reports or analyze exported data.
Examples of using OpenClaw alongside Saeree ERP:
- Report summarization — Export reports from Saeree ERP as PDF and have OpenClaw summarize the key points.
- Data preparation — Let OpenClaw help format data before importing it into Saeree ERP.
- Alerts — Set up a cron job for OpenClaw to check task statuses and send notifications via chat.
Warning: Never send confidential organizational data (such as customer data, financial information, or source code) to AI under any circumstances. Always use it in an isolated environment only. Read more about AI usage policies in the AI Governance article.
Conclusion
OpenClaw is a truly powerful AI Agent that enables ordinary people to have a "digital assistant" that actually works on their own machine — no coding required, just send commands via chat. However, this power comes with serious security risks that must be recognized before use.
| Suitable For | Not Suitable For |
|---|---|
| Personal experimentation (Personal Productivity) | Tasks involving confidential organizational data without isolation |
| Automation tasks not involving sensitive data | Deployment on production servers |
| Prototype / Proof of Concept | Organizations subject to strict compliance requirements |
| Developer tools / CI-CD automation | Using ClawHub skills without reviewing source code |
An AI Agent is like an extremely sharp knife — used skillfully, it can provide enormous help; used carelessly, it can cause serious harm. Organizations that want to use AI Agents like OpenClaw must invest in security infrastructure no less than they invest in the AI itself.
- Saeree ERP Development Team
If your organization needs an ERP system with audit trail and access control capabilities that support safe collaboration with AI, you can schedule a demo or contact our consulting team to learn more.
