We added Cursor agents as a coding-agent integration in Superconductor.
If your team already likes and uses Cursor, you can now bring Cursor agents into the same Superconductor workflow you use for the rest of your agents. Add the relevant key in Project Settings or Workspace Settings, then launch Cursor from the normal implementation flow alongside Claude Code, Codex, OpenCode, Grok Build, and the rest.
This also gives you more options for benchmarking agents on your own codebase. For example, one of our customers recently shared that they found Composer 2.5 was their best time vs. cost vs. quality tradeoff for simpler tasks after benchmarking it against the agents they were already using.
Available Models
The Cursor profiles include Composer 2.5, Composer 2.5 Fast, GPT-5.5 High, Claude Opus 4.8, Claude Sonnet 4.6, Gemini 3.1 Pro, Grok Build, Grok 4.3, and Auto.
Benchmark
We ran Cursor agents on a Personal SWE-Bench built from 12 ground-truth pull requests in Superconductor's Rails codebase.
In this run, Auto and Composer 2.5 are the highlighted Cursor profiles. On the time chart, they sit in the fast, high-quality cluster: close to the stronger-scoring agents, but finishing in roughly ten minutes instead of the twenty-plus minute range.
The cost chart tells a similar story from another angle. Auto is especially inexpensive here, while Composer 2.5 stays well below the most expensive agents without falling out of the competitive quality band.
You can run the same kind of comparison on your own repo. Pick merged pull requests your team is proud of, run your agent pool against inferred versions of those tickets, and compare quality, time, and cost on work that actually looks like your work.
Are there any other coding agents you would like to try out in Superconductor? Let us know.