Invite a friend, get 1 month of Pro freeHow it works

Every coding agent you use, in one window.

AgentVera is a desktop app that runs Claude Code, Codex, Gemini, Copilot, Cursor and 13 more coding agents side by side, per project. Each agent runs in a real terminal with your own install and your own account.

Start on the free plan; your first sign-in to the app turns on Pro for 7 days, no card. For macOS, Windows and Linux.

Architect Done auth-refactor 18k
✻ Claude Code opus · ~/code/acme-web
 
> Move session handling to refresh tokens
⏺ Let me look at the current session module first.
⏺ Read(src/auth/session.ts)
⎿ Read 142 lines
⏺ Update(src/auth/session.ts)
⎿ Updated with 42 additions and 17 removals
⏺ Bash(pnpm test auth)
⎿ ✓ 48 passed (1.9s)
⏺ Rotating refresh tokens now; reusing an old token will end the session.
⏺ Bash(git commit -m "auth: rotate refresh tokens")
⎿ [auth-refactor 3f2a91c] auth: rotate refresh tokens
Tester Done auth-refactor 9k
>_ OpenAI Codex (v0.155)
model: gpt-5-codex directory: ~/code/acme-web
 
› Write integration tests for this change
• I went through the existing auth tests.
└ Read session.ts, refresh.test.ts
• Ran pnpm vitest run api/auth
└ ✓ 12 passed
• Edited tests/api/refresh.test.ts (+86 -0)
─ Worked for 1m 04s ─
Frontend Done
✦ Gemini CLI · Using: 1 GEMINI.md file
 
> Fix the spacing on the settings page
✦ Looking at the settings components.
✓ ReadFile src/pages/Settings.tsx
✓ Edit src/pages/Settings.tsx
✦ Aligned the spacing to the 8 px rhythm.
Docs Waiting for you 11k
opencode · anthropic/claude-sonnet-4-5
 
┃ Update the README for the new API
← Read src/api/routes.ts
← Edit README.md +31 -9
△ Permission required: write docs/api.md
Allow once Allow always Reject
Dev server Working
~/code/acme-web (main) » pnpm dev
 
VITE v7.1.9 ready in 412 ms
➜ Local: http://localhost:5173/
➜ press h + enter to show help
12:04:31 [vite] hmr update /src/pages/Settings.tsx
6

Drag an agent here from the left

3 working 1 waiting for you
team@ 5s 12% Week 40% personal@ 5s 96% Week 71%

This is a demo: the real interface with pretend agents. Click around, build a flow, add an MCP server, start a new agent.

Agents write the code. AgentVera does the work in between.

Coding agents are strong, but running more than one is a job of its own. AgentVera takes that job: you watch every agent in one place, work moves from one agent to the next on its own, limits don’t stop it, and the code is checked before it reaches you.

Run several agents at once

Side-by-side panes · ⇧⌘A approval queue · pick up where you left off

Agents hand work to each other

Flows · team view · agent-to-agent messages · parallel team

Limits don’t stop the work

Account chains for Claude Code and Codex

More turns from the same subscription

Context profiles · automatic /compact · real token stats

Trust the code your agents write

Automatic code review · turns and undo · merge queue

Work moves on while you’re away

Night shift · automations · webhooks · Telegram

See what makes AgentVera different

The agents you already use. All of them.

AgentVera doesn’t bring a new model or a new account. It opens the CLIs on your machine and uses their logins as they are. Your subscription and API keys stay where they are.

18 coding agents and a plain terminal

Marked tools pick up AgentVera’s MCP servers and browser on every launch. Account chains and context profiles work with Claude Code and Codex.

Try adding an agent in the window

Every project gets its own desk.

Click a project and its agents open in the layout you left. Switch to another project and they keep working in the background.

  • Picks up where it left off

    Quit and reopen the app and every agent except plain terminals returns to the same conversation from its CLI’s history.

  • Keyboard first

    Move between agents with ⌘⌥ and the arrow keys, jump to a pane with ⌘1–9. Every shortcut can be remapped.

  • Queues and broadcast

    Queue a message while an agent is busy, or send one message to the agents you pick.

  • Real numbers

    Run time, agent hours, tokens and model mix come from your local conversation logs.

38 hagent hours this week

2.4 Mtokens

126prompts

Agent activity this week, by day and hour

When one agent finishes, the next one starts.

Wire agents together on a canvas. When the source agent finishes its turn, its last reply is dropped into a template and sent to the target. If the target is busy, the flow waits for its turn to end, and turn limits stop loops.

Yes No Start Claude Code Plan planner Condition reply contains “TODO” Codex Build Write tests for {{output}} Human approval Notify Telegram Yes No Start Claude Code Plan planner Condition reply contains “TODO” Codex Build Write tests for {{output}} Human approval Notify Telegram

Nodes: start, condition, wait, transform, notify and human approval.

Build one in the window
0 9 * * 1-5

Scheduled tasks

Send an agent a message on a cron schedule, like updating dependencies every weekday at 09:00. If the agent is busy, wait, send anyway or skip.

New in the latest release.

Your agents now work as a team: a team view with a manager, roles and hand-offs, CLIs that message each other, and agents that start their own parallel teams. Plus automations, a merge queue, tasks and Telegram.

  • Team view

    Turn your open agents into a team on an org chart: pick the manager, give each agent a role and instructions, and drag to draw who hands work to whom. While it runs, see each agent’s request, current step and time; answer permissions in place, stop an agent or redirect it.

    Group → Team

  • Agents start their own parallel teams

    Hand an agent a big job; it splits the work and gives each part to an agent in its own worktree. Finished parts merge into one integration branch and the check command runs; the outcome comes back to the agent as a message.

    Details → Parallel team

  • Every CLI talks to the others

    Claude Code, Codex, Gemini CLI, opencode and the other agents ask agents of their group questions, wait for replies and take work from a shared task list. In the team view the messages travel along the lines.

    agentvera-agents MCP

  • MCP gateway

    Reach remote MCP servers through AgentVera: the OAuth sign-in happens once in the app and every CLI can use SSE servers. Agents attach the servers they need to themselves or their group and ask you for new ones.

    MCP servers

  • AutomationsPro

    Let an agent start on its own when CI breaks, a pull request gets a comment or an issue is labeled. Set the rule once and choose which agent runs with which instructions.

    Details → Automations

  • Merge queuePro

    Agent branches land on the main branch one at a time, on top of its latest state. An agent resolves conflicts and a failing check gets two fix rounds; a branch that can’t be made green goes back, so main never breaks.

    Details → Merge queue

  • Tasks and reminders

    Keep your own to-dos with priority, dates and repeats. Signed in, tasks sync with your account and reminders arrive on time.

    Right panel → Tasks

  • Telegram with the AgentVera bot

    Connect to the AgentVera bot without setting up a bot of your own: see your agents, answer approvals, open a new agent and give work to a team or the night shift.

    Settings → Notifications

  • SSH vault

    Your SSH keys and server details sit in an end-to-end encrypted vault that nobody opens without your password. The vault syncs between your devices through your account.

    Settings → SSH vault

Hitting a limit doesn’t stop the work.

Line up as many Claude and Codex accounts as you like. When the remaining limit drops below your threshold, or the CLI reports a limit, the agent moves to the next account. The conversation is copied over and the agent carries on.

  1. personal@ Switched
    5-hour limit96%
  2. work@ Active
    5-hour limit41%
  3. spare@ Next
    5-hour limit8%

61% less context on the first request.

Pick a context profile per agent. That baseline is re-read on every turn, so the savings compound over long conversations. When context grows, AgentVera runs /compact at your threshold.

Claude Code

Standard 34.4k tokens
Balanced 23.7k tokens −31%
Lean 13.3k tokens −61%

Codex

Standard 13.3k tokens
Balanced —
Lean 9.7k tokens −27%

Context sent to the model on the first request of a new session. Measured with Claude Code 2.1 and Codex 0.155.

The agent finishes, a second model reads the code.

When a turn ends, the code it wrote, either the commits from that turn or all changes, is reviewed by the Claude model you choose. Findings come with a severity and a file:line link, and you can send them back to the agent in one click.

Up to two rounds in a row. You can also review the working tree or a single commit by hand.

Review3 commits from the auth-refactor turn
  • High

    Refresh token reuse isn’t detected; a new token is issued before the old one is revoked.

    src/auth/session.ts:88
  • Medium

    The database transaction isn’t rolled back on the error path.

    src/auth/store.ts:41
  • Low

    Unused import.

    src/auth/index.ts:3

No need to leave the terminal.

The tools your agents need live in the same app, each with its own shortcut.

  • MCP manager

    ⌘⌥M

    Playwright, GitHub, Sentry, Linear, Figma and more. Add from the catalog, test the connection, see the token cost of each tool. Secrets live in the Keychain.

  • Agent browser

    A separate Chrome profile. The agent browses with Playwright while you watch the tab live and step in when you want.

  • Git

    ⌘⌥G

    Changes, graph history and file diffs. Claude Haiku writes commit messages in your repo’s style.

  • Docker

    ⌘⌥D

    Containers, live logs, exec, images and compose, without opening Docker Desktop.

  • SSH and SFTP

    Saved servers, ProxyJump, port forwarding and two-pane file transfer. Run agents directly on the server.

  • Conversation search

    ⌘⇧F

    Full-text search across every Claude and Codex conversation, including ones started outside AgentVera. Resume any of them in a new agent.

  • Telegram

    Turn finished, errors, flows waiting for approval. Approve from your phone or reply to message the agent.

  • Command palette

    ⌘K

    Every action, agent, project and server in one search box.

Your code stays on your machine.

Agents run on your computer with your own CLI sessions. Your code and agent conversations are never sent to AgentVera servers; with team kanban, only the cards, comments and attachments you put on a board are shared.

  • Real terminals

    Every session is a real pseudo-terminal opened with node-pty.

  • No shell interpolation

    Commands start as an executable plus an argument array.

  • Keychain

    Passwords, tokens and MCP secrets live in your operating system’s keychain.

  • Sandboxed interface

    The UI has no Node or file system access; everything goes through typed IPC.

  • Scoped file panel

    It never leaves the project root and doesn’t follow symlinks.

  • Readable only by you

    Records are written atomically to a single file with 0600 permissions.

Start free, grow with your team.

Every core feature is on the free plan. Automation and team features such as flows, night shift, auto approval, the database manager and team kanban come with Pro and Team. Pro comes with a 7-day free trial.

Free

Free

Start free
  • Seats1
  • Agents2
  • Boards0
  • Storage100 MB
  • All core features: 18 agents, MCP, Git, Docker, SSH, code review

Pro

$9.99/ month

Try free for 7 days

Needs a card, but nothing is charged today. We email you 2 days before the trial ends; cancel by then and you pay nothing.

  • Seats1
  • AgentsUnlimited
  • Boards10
  • Storage5 GB
  • All core features: 18 agents, MCP, Git, Docker, SSH, code review
  • Flows
  • Team kanban
  • Agent database access
  • Agent SSH access
  • Database manager
  • Parallel team, task line and merge queue
  • Webhooks
  • Night shift
  • Auto approval
  • Runner on your own server
  • Settings sync across devices
  • Skill publishing
  • Automatic account switching

Team

$14.99/ month / seat

At least 2 seats

Subscribe
  • SeatsUnlimited
  • AgentsUnlimited
  • BoardsUnlimited
  • Storage50 GB
  • Custom domain✓
  • All core features: 18 agents, MCP, Git, Docker, SSH, code review
  • Live team view and handoff
  • Flows
  • Team kanban
  • Agent database access
  • Agent SSH access
  • Database manager
  • Parallel team, task line and merge queue
  • Webhooks
  • Night shift
  • Auto approval
  • Runner on your own server
  • Settings sync across devices
  • Skill publishing
  • Automatic account switching

A student or an open-source maintainer? A year of Pro is free: For students · For open-source maintainers

Prices are in USD and include VAT. Customers billed in Türkiye pay the TRY equivalent at the central bank (TCMB) selling rate on each charge day.

Questions

Is AgentVera an AI model?

No. AgentVera is a desktop app that manages the coding agents installed on your computer (Claude Code, Codex, Gemini CLI, Copilot, Cursor, Antigravity, opencode, Hermes, Qwen Code, Kimi, Kiro, Amp, Droid, Goose, Vibe, Crush, Aider and Cline) in one interface. Models and billing come from each agent’s own account.

Can I use a CLI that isn’t on the list?

Any CLI runs in a Terminal agent. The listed tools also take settings like model, effort and role from AgentVera.

Does it work with my existing subscription?

Yes. Each agent uses its CLI’s own login. If you signed in with a Claude or ChatGPT plan, AgentVera uses that session and doesn’t ask for an API key.

How much does it cost?

Core features are free, with 2 agents at once. Flows, night shift, auto approval, automatic account switching, the database manager and team kanban come with Pro and Team; current prices are on the Pricing page.

Which operating systems are supported?

AgentVera runs on macOS, Windows and Linux. Setup is the same everywhere: download, pick a project folder, open your agents.

Can I run an agent on a remote server?

Yes. When you create an agent, pick a saved SSH server and folder. The agent’s CLI, whether Claude Code, Codex or another listed agent, runs on the server with the server’s own CLI login and shows up in AgentVera like a local agent.

What happens to my agents when I quit?

Terminals are shut down cleanly. Next time you open the app, every agent except plain terminals returns to its conversation from its CLI’s local history.

Where is my data stored?

Project and agent records, your code and conversations stay on your computer (~/Library/Application Support/AgentVera on macOS); passwords live in your operating system’s keychain. When you use accounts, organizations and team kanban, your account details, device record and the cards, comments and attachments on boards are stored on the AgentVera server.

What does auto approval approve without asking?

On Pro and Team plans, for Claude Code agents: reading tools, read-only commands, test, lint, typecheck and build commands, and read-only git commands. Push, delete, irreversible git commands, output redirection and anything not on the lists come to you. Turn it off or change the lists in Settings → Auto approval.

Does my computer need to stay on for night shift?

Yes. Tasks run while the app is open and the computer is awake. Each task runs in its own worktree and branch, the agent doesn’t push, finished work is opened as a draft pull request and a report arrives in the morning.

Can I use it with my team?

Yes. Set up an organization with your AgentVera account and invite your team; on Pro and Team plans you get the team kanban. Agents still run on each developer’s own computer; the board shares cards, comments and assignments.

Do I need an account to use AgentVera?

Not for the core features: download, pick a project folder, open your agents. Pro and Team features such as flows, night shift, auto approval, the database manager and team kanban need you to sign in to the app with your AgentVera account.

Does dictation send my voice to a server?

No. Your voice is transcribed on your computer with the whisper.cpp build and model that ship with the app; nothing is sent anywhere.

What language is the app in?

Seven languages: English, Turkish, German, Spanish, French, Portuguese and Russian. New installs start in English; switch in Settings → Appearance.

Can I undo something an agent broke?

Yes. AgentVera snapshots the working folder to a hidden git ref at the start of every turn. From the Turns list you see the files a turn changed and undo it, without touching your branch or stash.

Which MCP servers can I use?

Add any of the 23 ready-made servers in the catalog (Playwright, GitHub, Sentry, Linear, Notion, Supabase, Figma, Canva, fal.ai, Meta Ads and more), or define your own over stdio, HTTP or SSE.

Bring your agents to one desk.

Download, pick a project, open your first agent. Your installed CLIs are ready to go.