Full coding agent
Claude Code
by Anthropic
Run Claude as a complete coding agent with Flame's project context, tools, permissions, and review flow.
- Sign in
- Claude login, API key, Bedrock, or Vertex
- Setup
- Installed on demand by Flame
AI Agents
Delegate the work. Keep control.
Flame's Agents Manager is a real control plane for AI work. Create, monitor, stop, and schedule tasks across projects with optional worktree isolation, scoped permissions, visible status, and an explicit landing policy.
Three full coding agents
Claude Code, Codex, and Gemini. One control plane.
Choose the agent that fits the task without giving up Flame's project context, worktree isolation, permissions, live activity, or review flow.
Full coding agent
Codex
by OpenAI
Use OpenAI's coding agent for full repository work while Flame keeps the task isolated and observable.
- Sign in
- Your existing Codex / ChatGPT login
- Setup
- Installed on demand by Flame
Full coding agent
Gemini via Antigravity
by Google
Run Gemini models as a full coding agent through Google's Antigravity CLI, with multi-turn context and tool activity in Flame.
- Sign in
- Your Google sign-in through Antigravity
- Setup
- Explicit install from Google's official CLI
Agents Manager DashboardLive status, tokens, and tool calls for every running, scheduled, and merging task, with a concurrency limit you set.
The Agents Manager
Six things you can do with every task
Not a chat box: a board for running, watching, and shipping AI work at scale.
Create
Start from a prompt, then choose the project, model, branch, isolation, and landing policy.
Monitor
Follow status, duration, token use, tool calls, and terminal output while each agent works.
Control
Stop, duplicate, restart, retry, or requeue a task without losing sight of its history.
Schedule
Run now, at a specific time, or after another job finishes. Chain dependent tasks into a pipeline.
Isolate
Give a task its own git worktree and branch so parallel agents stay away from your checkout and one another.
Review & land
Inspect the diff and merge preview, resolve blocked conflicts, then merge, squash, open a pull request, or push. Auto-merge stays opt-in.
How a task flows
From prompt to merged, safely
Every agent run follows the same reviewable path, so autonomy never means losing the plot.
Compose
Write a prompt, pick the project, model, branch, isolation, and how the work should land.
Isolate
Each task can run in its own git worktree and branch, so parallel agents do not need to share your checkout.
Work
The agent edits files, runs commands, and tests, with every action gated by your permissions.
Review
Inspect the diff and merge preview, ask questions, and decide how the work should land.
Land
Merge, squash, open a pull request, or push; automatic landing is available only when you opt in.
AI chat, in context
It already knows what you're looking at
- Bring in the context that matters: selected files, diffs, terminal output, test results, and live browser state.
- When the agent needs a decision, it asks: you answer with a tap, and it resumes.
- Tool cards show every edit, command, and test as it happens.
- Multi-project chat can grant scoped read or write access across several repositories.
- Switch model and effort per conversation, mid-thread.
Interactive Plan Mode
Plan first. Review in code. Execute with consent.
Toggle Plan Mode in AI Chat to enforce a server-owned planning gate. The agent investigates with read-only tools, drafts a Markdown plan directly in your editor, and waits for your edits and approval before touching a single file.
Read-only exploration
The agent inspects code, symbols, diagnostics, and browser state under a strict read-only sandbox. All file writes and shell execution are hard-denied.
Editable Markdown plan
Flame extracts the plan and opens it in a native, editable Markdown tab in the editor so you can review every file and step.
Iterative feedback
Directly edit the document in Monaco or click 'Request Changes' to send conversational feedback for another read-only planning turn.
Authoritative execution
Approval re-reads the exact edited Markdown file as the authoritative instruction, unlocks write tools, and runs in the same session context.
Hard read-only boundary:During planning and review, tool gating denies all write operations, command executions, and state mutations—regardless of saved permission rules. The review gate survives IDE restarts and tab switches so unapproved changes can never slip through.
Bring your own model
Cloud APIs, custom gateways, or fully local
Beyond the three full agent runtimes, connect Anthropic, OpenAI, Gemini, or OpenRouter APIs, route Claude through Bedrock or Vertex, or run local models through Ollama, LM Studio, and compatible endpoints.
Claude
Fable 5 · Opus · Sonnet · Haiku · Bedrock · Vertex
OpenAI API
Your OpenAI models
Gemini API
Google Gemini models
OpenRouter
Models exposed by OpenRouter
OllamaLocal
Any local model
LM StudioLocal
Any local model
Compatible endpoints
vLLM · LiteLLM · custom gateways
Beyond chat
AI woven through the whole editor
The same provider layer powers inline completion, next-edit prediction, local voice input, and focused AI actions, not just a side panel.
Voice
Local voice input
Use on-device Whisper transcription to dictate prompts and trigger supported command intents.
Control
Interactive Plan Mode
Groom, edit, and approve a structured implementation plan in a strict read-only sandbox before letting the agent write code.
Autocomplete
Inline completion
Fill-in-the-middle suggestions grounded in your project's symbols and nearby code.
Predict
Next-edit prediction
Cursor-style predictions of your most likely next change: jump and accept.
Actions
Focused AI actions
Fix a failed command, review a diff, repair failing tests, resolve conflicts, or improve a prompt from the surface where the issue appears.
Recap
Work summary
A standup-ready recap of commits, branches, project activity, time, CI state, and agent work.
Permissions & safety
Autonomy with a seatbelt
Every agent action is governed. You decide what's allowed, where, and whether to remember the choice.
Scopes
- Global
- Group
- Project
- This chat
Categories
- File read / write
- Terminal & git
- Browser
- IDE control
Decisions
- Allow once
- Deny
- Save a rule
- Ask every time
Guardrails
- IDE kill-switch
- Protected paths
- Command safety check
- Audit log (user vs agent)
Secret & Token GuardrailFlame warns immediately if sensitive tokens or credentials are detected in pending changes.
Put an agent on it.
Download Flame to delegate real work while keeping permissions, diffs, and landing decisions in view.
Free to use on macOS, Windows, and Linux.