Hey everyone, Version 1.10 is finally here, and it fundamentally changes how you interact with Agent Zero. What’s new? - The Universal Canvas: A brand new right-side panel where you can cowork with your agent. - Native Browser: Watch the agent browse the web live. You can even use the new Annotation Mode to visually select UI elements and tell the agent to fix them. - Office Magic: Work on Word, Excel, and PowerPoint files natively. The agent can even generate real Excel charts for you. - Time Travel: Messed up a file? The workspace now has a snapshot history. Diff, preview, and revert changes easily. Powered by Agent Zero's Space Agent. - OpenAI OAuth: Connect your existing OpenAI/Codex subscription so you don't have to burn API credits. Don't forget the CLI Connector 💻If you want the agent to work on your local machine's files while keeping its code execution safely inside Docker, install the CLI (requires v1.9+). - Mac/Linux: curl -LsSf https://cli.agent-zero.ai/install.sh | sh - Windows: irm https://cli.agent-zero.ai/install.ps1 | iex (Ask the agent to use the a0-setup-cli skill if you need help). Update Agent Zero:Settings > Update > Open Self Update > Restart and Update Go build something awesome. See you all tomorrow, April 29th at 3PM UTC in the Skool Community Call. Join us here: https://www.skool.com/live/DlyvNKHbyWw

Hey everyone, v1.9 just dropped, and it comes with a brand new tool: the A0 CLI Connector. What it does: It's a CLI that connects your local machine to your Agent Zero instance. It can view and edit files directly on your host machine, while all code execution stays safely inside Docker. Best of both worlds. The only requirement is installing Agent Zero version 1.9. Previous version don't support the CLI connector. Install it now: - Mac/Linux: curl -LsSf https://cli.agent-zero.ai/install.sh | sh - Windows irm https://cli.agent-zero.ai/install.ps1 | iex Or ask your Agent to use the a0-setup-cli Skill to guide you. 🔒 Security Note: This release also patches two CVEs (SSRF and Path Traversal). Please update your instances. Other highlights: - Browser Agent can now use its own dedicated model preset. - /project and /config commands work directly in Telegram, WhatsApp, and Email. - Redesigned messaging settings with clearer setup flows. Update Agent Zero: Settings > Update > Open Self Update > Restart and Update

Apologies if this has already been covered elsewhere. And if this does not apply, please do not hesitate to contradict this information. I'm heading on vacation and wanted this community to be aware. LiteLLM (used by Agent Zero since version 0.9.1 I believe) has become the victim of a supply chain attack that is impacting many people and organizations. Security breakdown on what happened can be found here here: https://www.youtube.com/watch?v=uwSjgv4otAk Key Details of the Integration - Unified Interface: LiteLLM allows Agent Zero to interact with over 100+ LLM providers (including OpenAI, Anthropic, Google, and local providers like Ollama or LM Studio) using a single OpenAI-compatible format. ⚠️ Critical Security Alert (March 2026) As of March 24, 2026, a major supply chain attack was identified affecting the LiteLLM library. If you are currently running or installing Agent Zero, please take the precautions outlined in the attached table. Verification Command: Run pip show litellm in your terminal. If you see version 1.82.7 or 1.82.8, treat your environment as compromised, rotate all API keys and secrets immediately, and delete your virtual environment.

Hey everyone, We just ran a stress test: Agent Zero vs. the new wave of closed-source agents. We threw a 28k row multi-table relational accounting dataset at a vanilla instance of Agent Zero running Grok 4.1 Fast Non Reasoning (via OpenRouter). Result: We completed a full P&L analysis, anomaly detection, and forecasting in under 60 seconds. That is roughly 5x faster than the closed competitors. You don't need a fancy, expensive subscription to get state-of-the-art analysis. You just need the right framework. Check out the video below. This is what happens when you combine a lightweight framework with a fast inference model. Who is going to try this with their own data?

Hi everyone! I am a cybersecurity professional and have been messing around with using Agent Zero as an automated penetration testing agent since Agent Zero was merged with the hacking edition. I am testing against the intentionally vulnerable OWASP Juice Shop web application which is also running in docker). After learning docker concepts, then Agent Zero concepts, I was able to push Agent Zero pretty far in solving Juice Shop challenges. However, the sequential nature of Juice Shop challenge solving was taking a long time and thus I started looking into using subordinate agents for each challenge (with custom profiles for each, pentestmanager, reconspecialist, pentester). Alas, Agent Zero is not architected to support simultaneous agent swarms at this point. I then pivoted to a custom single agent using the Task Scheduler to create simultaneously executing tasks (using TaskType.AD_HOC, and then run_task with dedicated_context=true) and this appears to be working well so far. I also went to through the journey of attempting to use local LLMs via LM Studio before finally listening to folks here and instead using free LLMs on OpenRouter (currently using Chimera). I still hold hope that I can self-host a ridiculously large opensource LLM, in a private cloud if needed. Abliterated LLMs are particularly useful for ethical hacking and local LLMs are useful for sandboxing cyberlab activities when commercial LLM decline actions. However, there does not appear to be an Agent Zero setting (evironment variable or setting dialog) to adjust LLM timeout value, and slow LLM responses invariably lead to timeout errors. It would be fun to develop the solution to the point where it can target any cyber range system (e.g. infrastructure, edge device) and the agent follows industry penetration testing framework stages and actions from PTES (Penetration Testing Execution Standard), NIST SP 800-115, and OSSTMM, etc.