FAQ

Questions, answered

The things people ask most about Flame. Can't find it? Reach out, we read everything.

General

What is Flame?
Flame is an AI-native IDE: a full development workspace for people who juggle many projects at once. It pairs the editor you know with multi-project workspaces, git worktrees, an integrated and broadcast terminal, live previews, and an Agents Manager for AI tasks.
How is it different from VS Code?
Both editors support multiple folders, Git worktrees, and agent workflows. Flame emphasizes project discovery and groups, broadcast commands, integrated developer tools, and a shared task manager. VS Code has a much broader extension ecosystem. Compare your actual project, required extensions, and daily workflow on our comparison page.
Which languages are supported?
Bundled tooling covers TypeScript/JavaScript, HTML, CSS/SCSS/LESS, JSON, YAML, Bash, Dockerfile, PHP, Vue, Svelte, Astro, and Markdown. Flame also drives language servers for Python, Rust, Go, C/C++, Java, C#, Kotlin, Swift, Dart, and PowerShell when they are available on PATH. Linters attach when the required tool or project configuration is present.
What platforms does Flame run on?
Desktop builds target macOS 13 (Ventura) or later on Apple Silicon and Intel, Windows x64 and ARM64 installers, and Linux as an x64 or arm64 AppImage plus an x64 .deb package.
Does Flame include debugging and profiling?
Yes. Flame supports breakpoints, stepping, variables, watches, and call stacks over the Debug Adapter Protocol. When you use Flame's built-in Node debug engine, the profiler can record interactive CPU flame graphs, mark hot source lines, capture heap snapshots, and compare allocations.
Can I extend and customize Flame?
Yes. Flame can install and manage local manifest extensions with declarative contributions for languages, snippets, themes, formatters, and linters. You can also customize themes, density, shortcuts, gestures, snippets, formatters, and project or group run configurations, then move portable configuration between installations as scrubbed JSON.
Which test frameworks can Flame run?
The Test Explorer has built-in adapters for Vitest, Jest, Node's test runner, Pytest, Go test, and Cargo test. It supports discovery, run and watch workflows, grouped results, and coverage gutters where coverage data is available.

AI & agents

Which AI agents and models can I use?
Flame runs Claude Code, OpenAI Codex, and Google Gemini through the Antigravity CLI as full coding agents, using your existing Claude, ChatGPT, or Google sign-in. AI Chat also supports GitHub Copilot. You can connect Anthropic, OpenAI, Gemini, OpenRouter, Bedrock, Vertex, Ollama, LM Studio, vLLM, LiteLLM, or another compatible endpoint for chat, inline completion, and focused AI actions, with defaults per surface and model overrides per conversation.
Can a single conversation delegate to other agents?
Supported native Claude sessions offer Multi-agent mode for read-only research and review workers. The lead retains edits and commands. Claude and Copilot worker activity can be inspected when the configured runtime exposes it. Codex, Antigravity, and local endpoints do not all share these native capabilities; separate worktree jobs remain a different workflow.
Can I import conversations from another coding tool?
Yes. Settings → AI → Import and export chats supports selected Claude Code, Codex, Copilot, Claude and ChatGPT exports, and Flame archives. Preview histories, select conversations, and map them to projects or global chats. Importing a history does not restart its workers.
Can agents run while I keep working?
Yes. Each task can run in its own git worktree and branch, so multiple agents work in parallel without touching your working tree, and you set a concurrency limit for the whole fleet.
How do agents avoid breaking my code?
You choose each job's project, permissions, branch strategy, and landing policy. A task can run in an isolated worktree, while tool activity, diffs, test results, and conflicts stay visible. Manual review is the default; automatic landing is opt-in.
What is AI Agent Plan Mode and how does it protect my codebase?
Plan Mode locks the agent into a strict, server-enforced read-only planning phase where it can inspect files, search symbols, and read diagnostics, but write tools and shell execution remain disabled. It produces a Markdown plan that opens in a dedicated Plan tab: decisions with each option's consequence and a recommendation, an at-a-glance summary, an optional architecture diagram, section feedback, and saved versions. Ask AI about a section, request a revision in another read-only turn, or edit the Markdown; write tools are unlocked only when you choose Approve & Implement.
Can I use Flame fully offline?
The core editor, local files, Git, and terminals work offline when their required tools are installed. AI requests can stay on your computer when you select a locally hosted model through Ollama, LM Studio, or another compatible endpoint. Cloud agents, integrations, model downloads, and mobile relay connectivity require a network connection. The packaged app also checks for software updates.

Privacy & security

Does my code leave my machine?
Local desktop projects are read on your computer, with the default backend on 127.0.0.1. When you use cloud AI, prompts, selected context, and agent tool results can be sent to your configured provider. This can include source code, terminal output, and browser data. A model hosted on your computer keeps model requests local. SSH projects are processed on the remote host. Read the privacy policy for each connection type.
Where are my API keys stored?
Flame-managed API keys and integration tokens are encrypted at rest. Packaged desktop builds use the operating system's credential protection when available. Standalone or headless servers can fall back to a local key file protected by filesystem permissions. Signed-in agent runtimes also manage credentials through their own authentication systems.
Is there an audit trail?
Yes. Security-relevant actions (opening a terminal, writing a file, pushing to a remote, and every tool an agent invokes) are recorded to an audit log that distinguishes you from the agent.
Does Flame send telemetry?
Usage statistics are optional and off by default. With your consent, Flame sends independent, identifier-free software and feature reports to its self-hosted service. The revised notice also covers pseudonymous installation activity, which remains unavailable pending privacy release review and publication. You can preview the payloads and manage or delete linked usage data in Settings → Diagnostics. Crash reports are separate and are sent only when you choose Send.
How does Secured Access protect my workspace?
Secured Access keeps the Flame server on loopback while the host dials out through an end-to-end encrypted, content-blind relay. The relay cannot decrypt workspace traffic, but it can observe limited connection metadata such as timing, byte volume, and live socket IP addresses. In Remote Access, choose Secured Access, generate a short-lived QR code, and scan it inside the native Flame Mobile app. Enter the phone's two-digit confirmation code on the desktop, then choose a device access duration. Scanning alone does not grant access. Switching access Off, revoking a device, or expiry ends its active sessions. The desktop must be running and connected. Flame Mobile for iOS and Android is available soon. See the pairing guide.

Access & pricing

What is available now, and what is coming soon?
Flame IDE desktop downloads, the Chrome Browser Bridge, and the desktop Mobile Preview panel are available now. Mobile Preview uses your Android tooling or macOS with Xcode for iOS simulators. The separate native Flame Mobile companion for iOS and Android is available soon. Its desktop Secured Access pairing support is implemented, but a desktop download does not install the phone app. Check desktop downloads and the mobile availability page.
How do I get Flame?
Flame is available directly on our download page for macOS 13 (Ventura) or later, Windows, and Linux. Choose your operating system and start building.
What will it cost?
Flame is free to use. Cloud AI providers and other connected services may require their own subscription or usage fees. You can also configure a local model endpoint.

Still curious?

Download Flame and see for yourself, or ask us anything from the contact page.

Free to use on macOS 13 (Ventura) or later, Windows, and Linux.