How much time will I actually save with Fazm?

Most users report saving 45-90 minutes per day on repetitive tasks. The biggest time savings come from email management (drafting replies, sorting, archiving), form filling (expense reports, CRM updates, applications), research (comparing products, gathering data from multiple sites), and social media (cross-posting, scheduling, replying). For example, booking a flight that normally takes 10 minutes of clicking through airline sites takes a single voice command with Fazm. Filing an expense report that requires opening a PDF, extracting numbers, and entering them into a spreadsheet becomes one spoken sentence. The more repetitive and multi-step your daily work is, the more time you get back.

What exactly can Fazm control?

Fazm controls everything you can do manually on your Mac. That includes mouse movements and clicks, keyboard typing, browser navigation and form filling, app switching, file management in Finder, and system-level actions. It uses direct browser DOM control for web interactions - meaning it reads and manipulates the actual page structure rather than taking screenshots and guessing where to click. For native macOS apps, Fazm uses accessibility APIs to interact with buttons, menus, text fields, and other UI elements. This works across any application: Chrome, Safari, VS Code, Slack, Figma, Terminal, Mail, Calendar, and more.

How is Fazm different from ChatGPT or Claude?

Chatbots like ChatGPT and Claude give you answers in a conversation window. Fazm takes action directly on your computer. When you say 'reply to Sarah's email,' Fazm opens your email app, finds Sarah's message, writes a contextual reply, and sends it. A chatbot would tell you what to write, but you still have to do all the clicking and typing yourself. Fazm moves your mouse, types on your keyboard, clicks buttons, and navigates between apps. It is a computer agent, not a chat assistant.

How does the memory layer work?

Fazm builds a personal knowledge graph by extracting information from your files, browsing history, conversations, and daily activity. Over time, it learns your contacts (names, emails, relationships), accounts and credentials, preferences (formatting, tone, scheduling habits), and frequently used workflows. In your first week, you might say 'Reply to Sarah's email, she's my cofounder, her email is sarah@acme.com.' By week four, you just say 'Reply to Sarah' because Fazm already knows who she is. By week eight, Fazm can draft the reply before you even ask. All of this data stays locally on your Mac - your knowledge graph never leaves your machine or gets sent to any cloud service.

Is Fazm safe? Can it mess up my computer?

Safety is built into every layer of Fazm. You watch every action in real-time on your screen - Fazm moves your mouse and types visibly, so there is nothing hidden. You can stop any action at any point by pressing a keyboard shortcut. Fazm also shows you what it plans to do before executing destructive actions like deleting files or sending emails, giving you a chance to confirm or cancel. Because Fazm is open source, the code is fully auditable on GitHub. There is no hidden behavior, no background processes making changes you cannot see, and no actions that bypass your screen.

What platforms does Fazm support?

Fazm currently supports macOS, running natively on both Apple Silicon (M1, M2, M3, M4) and Intel Macs. It is built with Swift and uses macOS-specific frameworks like ScreenCaptureKit for screen analysis and Accessibility APIs for app control, which is why macOS is the primary platform. Windows and Linux support are on the roadmap but do not have release dates yet. You can also control your Mac remotely from your phone using Fazm Remote, which gives you voice access to your desktop while you are away from your computer.

How does Fazm voice input work?

One keyboard shortcut activates push-to-talk. Hold the key, speak naturally about what you need done, and release. Fazm transcribes your speech locally using on-device models, so there is no network latency for the voice recognition step. You do not need wake words like 'Hey Siri' or 'OK Google' - just press and talk. You can also type commands into the floating toolbar if you prefer text input. Fazm understands natural language, so you do not need to memorize specific commands. Say 'book me a flight to Tokyo next Thursday' or 'find the cheapest AirPods on Amazon' and Fazm figures out the steps.

Is my data sent to the cloud?

Fazm processes screen analysis and memory entirely on your machine. The personal knowledge graph, file indexing, and screen capture all run locally. The only data that leaves your Mac is the intent - a text description of what you want to accomplish, which is sent to the AI model (Claude or GPT) for action planning. Your actual screen content, file contents, browsing history, and personal knowledge graph never leave your computer. Fazm is also open source, so you can verify exactly what data is transmitted by reading the code on GitHub. For users who need complete air-gapped operation, local model support is planned.

Optimizing Multi-Step Agents - Keeping a Running Log to Prevent Action Loops

Fazm Team·March 17, 2026·3 min read

multi-step-agents action-loops running-log agent-optimization debugging

The Loop Problem in Multi-Step Agents

If you've built a multi-step AI agent, you've probably seen this: the agent completes step 3, then goes back and does step 2 again. Or it checks the same file three times. Or it runs the same terminal command in a loop, each time acting like it's the first attempt.

This is the action loop problem, and it kills both token budgets and user trust.

Why Agents Loop

LLMs don't have built-in state tracking. Each decision is based on the current context window, and if the context doesn't clearly show what's already been done, the model has no reason to skip a step. It sees "file needs to be updated" in the instructions and updates it again, even if it already did.

Long-running tasks make this worse. As the context fills up with tool outputs and intermediate results, earlier actions get pushed further from the model's attention. The agent literally forgets what it did ten minutes ago.

The Running Log Fix

The solution is straightforward: maintain a running log of completed steps and inject it into the agent's context before each decision point.

The log doesn't need to be fancy. A simple list works:

Completed: Checked git status
Completed: Created feature branch
Completed: Updated config.json
Completed: Ran test suite - all passing
Current: Preparing commit message

This gives the agent a clear picture of where it is in the workflow. It won't re-run tests if it can see they already passed. It won't recreate a branch that already exists.

Implementation Details

Write the log to a file that persists across context resets. If the agent's context window fills up and gets trimmed, the log file survives. The agent can read it fresh and pick up exactly where it left off.

Keep log entries concise but specific. "Updated file" is too vague - the agent might think a different file needs updating. "Updated src/config.json - added API endpoint for staging" is clear enough to prevent repeats.

The Broader Lesson

Most agent reliability problems come from poor state management, not model capability. The model is smart enough to do the work. It just needs to know what work has already been done.

Optimizing Multi-Step Agents - Keeping a Running Log to Prevent Action Loops

The Loop Problem in Multi-Step Agents

Why Agents Loop

The Running Log Fix

Implementation Details

The Broader Lesson

More on This Topic

Related Posts

Don't Trust Agent Self-Reports - Verify with Screenshots

The Irony of AI Automation - Debugging Skills Takes Longer Than the Original Task

Error Handling in Production AI Agents - Why One Try-Except Is Never Enough