Open Free Max docs
Everything OFM can do — from a single agent tab to a full Mission Control of parallel sessions. Drive Claude Code, Codex, Antigravity, OpenCode, Mistral Vibe, Kimi Code and Cursor Agent interactively, on your subscription.
Getting started
Introduction
What Open Free Max is, who it's for, and how it stays light, durable and on your subscription.
Install
Download and install Open Free Max on Windows or Apple Silicon macOS.
Quickstart
Open a project, start your first agent session, send a prompt, and discover Mission Control in five minutes.
Pricing & plans
What Open Free Max costs — the OFM Pro plan and the 7-day free trial with no credit card. OFM Pro is the IDE licence and is separate from your agent provider subscription.
Account & billing
Create an account, start the 7-day free trial without a card, manage your subscription and invoices — and understand why the OFM plan is separate from your AI provider's plan.
Concepts
How OFM works
The architecture behind Open Free Max — a native Tauri shell that drives official CLIs through an embedded PTY, with a small local Rust core.
Subscription vs API
Why Open Free Max keeps your agent usage on your Max/Pro subscription and never on the metered API bill — and exactly what it refuses to do.
Supported agents
The agent CLIs Open Free Max drives — Claude Code, Codex, Antigravity, Cursor Agent, OpenCode, Vibe and Kimi Code — and how provider support is structured.
Memory & Obsidian
Understand OFM's persistent project memory, review what your agents learn, and connect a live Obsidian vault through MCP.
Agent CLIs
Supported agent CLIs
Open Free Max drives seven official agent CLIs — Claude Code, Codex, OpenCode, Mistral Vibe, Antigravity, Kimi Code and Cursor Agent — in real interactive mode, and this table shows exactly what each one supports.
Claude Code
Claude Code is the most fully supported CLI in Open Free Max — exact session resume, MCP server injection, project memory as a tool, and exact token accounting.
Codex
How Open Free Max runs the OpenAI Codex CLI — resume by id, sandbox and approval policy, and MCP servers injected as command-line overrides that never touch your config.toml.
Antigravity
Google's Antigravity CLI (agy) in Open Free Max — model choice beyond Gemini, conversation resume by id, and MCP wiring through a project-scoped plugin instead of your global config.
OpenCode
Running the OpenCode CLI inside Open Free Max — model selection in provider/model form, agent profiles, the embedded server port, and how permissions are driven by environment rather than flags.
Mistral Vibe
Mistral's Vibe CLI in Open Free Max — devstral models, approval profiles from plan to auto-approve, and full history ingestion into project memory.
Kimi Code
Moonshot's Kimi Code CLI in Open Free Max — plan, auto and yolo permission modes, Skills folders, and the Git Bash requirement on Windows.
Cursor Agent
Cursor's headless-free agent CLI runs in Open Free Max as a plain interactive session — what works today, and what the CLI does not expose.
Using the IDE
Projects & workspaces
Open, clone and organize projects in Open Free Max — the left rail, recents, and drag-and-drop reordering shared with Mission Control.
Native WSL2
Open projects that live inside a WSL2 distro and run agents in Linux from Open Free Max, with the right PATH and no IDE-side install.
Sessions & persistence
Run multiple agent sessions per project and never lose a conversation — Open Free Max resumes each tab from the CLI's own durable store.
Screenshots
Paste or drag any number of images into a prompt — Open Free Max saves them, shows thumbnails, and injects real file paths at send time. macOS system screenshots are imported automatically.
Agent settings
Configure the launch flags of every agent CLI from one schema-driven form — model, permission mode, sandbox, extra folders, environment variables — applied at the start of each session.
Explorer & editor
Browse and search the project tree, open and edit files, preview Markdown — a lightweight file panel that sits next to your agents instead of competing with them.
Integrated terminal
A real shell in the same folder as your agents — several tabs, stacked under the agent panel, so you can run a build or a git command without leaving the workspace.
Prompt shortcuts
Ten numbered phrases you send to the active agent with one key — "commit and push", "run the tests", "keep going" — global or scoped to a single project.
Mission Control
Mission Control
Supervise every running agent on one screen — the live wall, agent states, and the unified confirmation queue.
The answer queue
Every agent waiting for a confirmation, in one list you clear with single keystrokes — approve, reject, or approve and remember so the same question never comes back.
Auto-approval & guardrails
Let agents keep moving with a rules engine, while a hard deny-list, audit trail and kill-switch keep you safe — at any autonomy level.
Launching sessions
Start an agent in any project from the Dashboard — pick the CLI, write the first prompt, attach recent screenshots or files, and let it send itself as soon as the agent is ready.
Audit log
Every agent confirmation that got an answer, with the source that decided it — you, a rule, the autonomy level or the deny-list — so you can find and fix whatever let an agent act on its own.
Mission Control settings
Choose who answers your agents' questions — globally or per project — plus dangerous-action blocking, activity detection tuning, notifications, hotkeys and the emergency stop.
AI Router
AI Router
The AI Router reads your brief, picks the level of thinking the work needs, picks which of your subscriptions runs it, and shows its choice before anything launches.
Levels and capabilities
The three levels the AI Router works with, how to name one yourself, and what it tells you when a task needs something in particular.
Quota and subscription switching
How OFM watches each subscription's quota window, keeps a reserve it never spends, and moves unattended work to another subscription you have already used on that project.
Several opinions at once
Tick more than one subscription and the same task runs on each, every agent with its own model — and what the router remembers, per project, from the times you disagree with it.
Recurring work
Recurring work
Wake an agent on a schedule, hand it a checklist, and let it report only when there is something to report — OFM's recurring jobs, on your own subscription.
Checklists and reports
How to write the checklist a recurring job reads every turn, and how the report-or-acknowledge convention keeps notifications meaningful.
Scheduling
Intervals, fixed daily times, active-hours windows, sleep behaviour, and every state a recurring job can display.
Approval mode and CLI flags
The setting that decides whether unattended turns work at all — per-job approval flags for each CLI, and why programmatic flags are refused.
Quota, limits and safety
What a recurring turn consumes, the caps you can raise or lift, and exactly what an unattended agent can overwrite in your project.
Memory & knowledge
Project memory
Open Free Max keeps a local, per-project memory — every past agent conversation indexed for search, plus human-validated facts your agents can read at launch. Nothing leaves your machine.
History search
Search everything your agents have already done on a project — across Claude Code, Codex, Antigravity and Vibe — and reopen the original conversation from a result.
Project knowledge
Turn raw agent history into a small set of durable, human-validated facts — decisions, constraints and conventions — that every future session reads before starting work.
Memory injection
How validated facts and searchable history actually reach an agent — an MCP tool it can query on demand, and a generated block in the project's guidance file at launch.
Integrations
MCP servers
Declare the MCP servers your agents should get — a local stdio program or an HTTP service — scoped to one project or to all of them, and wired in at session start without touching your CLI configuration.
Connect an Obsidian vault
Give your agents read and write access to a live Obsidian vault through the Local REST API plugin — set up in four fields, with the key never leaving your machine.
Usage & costs
Consumption & activity
See how much each CLI and model actually consumed on a project — messages, input, output and cache tokens — plus session and agent activity, measured locally with no API call.
Savings vs API pricing
Open Free Max prices the work your agents did against public per-model API rates and compares it with what your subscription costs — the concrete number behind "subscription, never the meter".
Telemetry & privacy
Exactly what Open Free Max sends, what it never sends, and how to turn it off — pseudonymous product telemetry grouped by a random install id, with no content, no paths and no stored IP.
Reference
Settings
Every Open Free Max setting — per-provider launch flags, Mission Control behavior, theme and language.
Keyboard & prompt shortcuts
Keyboard shortcuts in Open Free Max, plus reusable prompt phrases you can trigger in any project or send directly to an agent.
Data & file locations
Where Open Free Max stores its settings, memory, usage and Mission Control data — and why your agent transcripts live elsewhere.
Troubleshooting
Fixes for common Open Free Max issues — CLI not found, installer checks, agents that don't start, sessions that stall, and what to report after a crash.
FAQ
Frequently asked questions about Open Free Max — billing, API keys, platforms, local memory, and how it compares to Electron IDEs.