Agents
Choose which coding agents Proliferate can run for your team.
Proliferate runs your team's coding agents inside the same governed workspaces: every agent gets repo context, terminals, and a review surface, and you keep control over credentials and access.
Pick the agent that fits the task. Proliferate maintains its own copy of each
agent, pinned to an exact version, and reconciles it every time the runtime
starts — installing anything missing and reinstalling anything that has
drifted — so there's nothing to set up ahead of time on a new machine. An
agent you installed yourself on PATH is left untouched, and the managed
copy takes precedence over it.
Supported agents
Proliferate runs Claude Code, Codex, Cursor, OpenCode, and Grok. Each gets its own harness page under Settings -> Agents, where you pick its authentication route and the models it may use.
Models
The model picker reflects the models currently available through the selected agent and its authentication route — native sign-in, your own API key, and the gateway each expose a different set. There's no version to pin or upgrade yourself.
| Agent | Models |
|---|---|
| Claude Code | Anthropic's current model lineup. |
| Codex | OpenAI's current model lineup. |
| Grok | xAI's current model lineup. |
| Cursor | Cursor's own current model lineup, managed through its own account. |
| OpenCode | Whichever provider you connect, from Anthropic and OpenAI to a hundred-plus other options. |
Browse exactly what's reachable right now under Settings -> Agents -> an agent -> All Models.
Authentication
Depending on the agent, model access comes from its native local sign-in, an API key you bring, or the Proliferate gateway. Cloud workspaces and their separate authentication surface are coming soon.
Admin controls
Next: set up a workspace to run an agent in.