Cursor Agent provider

The Cursor Agent provider hands each agent to Cursor's own agent. Like Codex, Cursor runs the work in its own engine with its own tools, while Cerevisor supervises: the order agents run in and what is handed from one to the next stay with Cerevisor.

Why use it: you have a Cursor account and want Cursor's agent in your workflows, with optional cloud execution that survives Cerevisor restarts.

Prerequisites

• A Cursor account with an API key from cursor.com/dashboard.
• Nothing to install by hand. If Cerevisor tells you this copy of the app is missing the part that talks to Cursor, reinstall Cerevisor from cerevisor.com — your workflows and settings are kept.

Setup

1. Answer An API key in the setup pane, then click the Cursor tile — or Settings → Providers → + Add provider.

Three steps:

1. Check

Cerevisor confirms the part of the app that talks to Cursor is present. If it isn't, it says so in one sentence and points at a reinstall.

2. API key + validate

Paste your Cursor API key. Click Validate — Cerevisor asks Cursor to confirm it.

3. Connection check

Cerevisor asks Cursor for one trivial answer. It counts as a pass only when Cursor reports the work finished and sends real text back — an empty or unfinished reply is a failure, not a pass.

Save and you're done.

Local vs. cloud

Cursor supports two execution modes:

Mode	Where it runs	Use when
Local	On your machine.	Most cases. Standard agent run, completes when Cerevisor is running.
Cloud	On Cursor's VMs.	Long-running runs. Survives Cerevisor restarts and laptop sleep.

Toggle the run mode per workflow in Settings → Workflow → Run mode (Local / Cloud). Cloud mode only applies when the active provider is Cursor Agent. See Cursor cloud runs for the full cloud setup.

Permissions

Cerevisor sets Cursor up from your workflow's permissions:

Your workflow's permissions	What Cursor gets
File writes or shell commands allowed	Cursor may make changes.
Both blocked	Cursor is kept read-only.

As with Codex CLI, what Cursor may do is set before it starts: you are not asked about individual actions during the run.

Known v1 limitations

• No mid-run continuation. A Cursor run can't be steered or continued mid-task; when Cerevisor's output check finds a problem, the retry runs as a fresh attempt. Cloud runs can be chatted with after they finish (the run persists on Cursor's servers and resumes with full context); local runs can't — use "Resume from this agent" to re-run instead.
• Cost reported as $0.00. Cursor doesn't report usage for local runs, so Cerevisor can't compute spend even though Cursor bills your account per use.
• MCP tools. MCP servers you added in Cerevisor are not passed to Cursor runs today. If you need MCP tools inside a Cursor agent, set them up in Cursor directly.

Background features

Background features (run retrospectives, memory upkeep, skill summaries) still work when Cursor is your default: each call runs as one small Cursor agent task. If a call fails, Cerevisor automatically retries it on another provider in your library that can answer it and records any fallback spend in Analytics → Models.

The chat builder works two ways:

• Point it at a chat-capable provider (most reliable) — Settings → Chat Builder lets the builder run on a different provider than your agents.
• Run it on Cursor directly — turn on Settings → Chat Builder → "Build with a subscription provider." Each builder message becomes one Cursor agent task that emits the workflow changes. Slower and less precise than a chat-capable provider, and it uses your Cursor account, but it keeps everything on one provider.

When to prefer Cursor over Codex

• You're already a Cursor user.
• You want cloud execution that survives Cerevisor restarts (Cursor has this; Codex doesn't).

When to prefer Codex over Cursor

• You're a heavy ChatGPT user with a subscription you want to amortize.
• You don't want to pay per token (Codex's subscription model vs. Cursor's per-token).
• You prefer OpenAI's models over Cursor's curation.