VibeAround v0.6.3: a web workspace for multiple coding agents
VibeAround v0.6.3 is out.
This release wraps up the bigger v0.6.x shift: VibeAround is becoming a hub for coding agents across desktop, web, terminal, and messaging.
The biggest visible change since v0.5.x is the Web experience: it is now a web-based workspace for running different coding agents side by side, one place to launch, resume, inspect, and switch between agents.
From the browser, you can now:
- Start Claude Code, Codex CLI, Gemini CLI, OpenCode, and other agents
- Resume previous sessions
- Switch between agents and provider profiles
- Review code changes with rendered diff views
- Attach files, images, and documents
- Control thinking/tool visibility
- Open Web Terminal for local PTY sessions
- Launch live previews for dev servers, Markdown, and HTML
v0.6.3 also cleans up the model-routing layer:
- Renamed the old API proxy wording to API Bridge across the app, docs, profile routing, and launch UI
- Published the bridge layer as an independent Rust crate: va-ai-api-bridge, nicknamed va-aab
- VibeAround now consumes va-ai-api-bridge from crates.io instead of a submodule/local checkout
- Bridge routes remain compatible with OpenAI Responses, OpenAI Chat Completions, Anthropic Messages, and Gemini Generate Content
- Improved IM command workflow controls
- Refreshed dependencies and release packaging
Release:
https://github.com/jazzenchen/VibeAround/releases/tag/v0.6.3
Bridge crate:
https://crates.io/crates/va-ai-api-bridge
Feedback is very welcome, especially from people using multiple coding agents or non-default model providers.
Replies
@jazzenchen A browser-based coding hub sounds handy, why did you choose web instead of desktop?
VibeAround
@tessa_lynch Great question. The idea is not to replace the desktop app, but to separate responsibilities better.
I want the web workspace to become a unified entry point for heterogeneous coding agents, so users do not have to keep switching between different tools and UIs. The desktop app can stay focused on local runtime, permissions, bridge, and tunnel capabilities.
Another big reason is mobility. Since VibeAround already has built-in tunnel support, the web experience can later be accessed from other devices naturally, without asking people to install another mobile app.
The side by side agent workflow is probably the most interesting part here, managing multiple coding agents across different interfaces still feels fragmented in most tools . Nice direction for simplifying that experience.
VibeAround
@jared_coleman Exactly. That fragmentation is the main pain I am trying to reduce.
There are already many good coding agents, but they often live in separate terminals, desktop apps, or provider-specific flows. VibeAround's web workspace is meant to give them one shared operating surface: launch, resume, inspect, preview, and switch without constantly changing tools.
Side-by-side agents is a strong idea, does it help compare results easily?
VibeAround
@sanjau_sanjay Yes, that is definitely where I want to go.
Today, side-by-side agents already make it easier to inspect different sessions, diffs, and previews in one place. A more explicit comparison workflow is on the roadmap.
The direction I am excited about is comparing freely across model and agent combinations: different models in the same agent, the same model across different agents, or different provider profiles running the same task.
Curious how isolated the agents are from each other internally. Can they share context or is separation intentional?
VibeAround
@owen_parker4 Separation is intentional for now.
I am pretty cautious about forcing agents to share context implicitly, because I do not want to disturb each model's own session history or make debugging harder. Today, agents can still share through the workspace and files.
That said, I have a promising idea here: using ACP for explicit history fork flows, so context sharing can become deliberate instead of magical. Stay tuned.
Feels like the real value here is reducing friction between tools rather than replacing any individual coding agent.
VibeAround
@dustin_warren Exactly. I am not trying to replace Claude Code, Codex CLI, Gemini CLI, OpenCode, or any other good agent.
The value I care about is reducing switching costs between them. VibeAround should be the place where you launch, resume, inspect changes, preview output, and route models or providers through a single surface, while each agent can still keep its own strengths.
The rendered diff reviews and live previews are a huge quality of life upgrade. Makes the workflow feel closer to a real IDE experience.
bem interessante