Superconductor
79 posts
Superconductor
@Superconductor
The multiplayer cloud workspace for your team and AI coding agents.
Joined June 2025
- Superconductor reposted
Good stuff: we launched Fable 5 on a real memory leak ticket alongside GPT-5.5 and Opus 4.8 (a total of 5 agents in one click). Fable was the ONLY ONE that found the root cause: an event listener dependency holding refs to replaced DOM nodes. Red before. Green after. Hell yeah.
Coding agents will do whatever they think it takes to get the job done -- including accessing malicious domains. Network Sandboxing gives teams centralized control over which domains coding agents can access. Block everything, allow specific domains, or enable full access.
00:00- Let agents read and update your @Jira tickets with Atlassian Rovo MCP on Superconductor. Look up issues, comment, summarize work, create follow-ups, update fields, and move issues through workflows.
- Superconductor repostedWe ran the new Grok Build harness + model on our "Personal SWE-Bench" for our Ruby on Rails codebase. The results are better than expected! Fast and pretty good. We're now throwing Grok Build into the mix when we launch multiple implementations for tricky features.
- AI coding agents produce a lot of code quickly. Superconductor's QA Review helps you decide what deserves your attention before you dig into the diff. It checks functionality and code quality, whether tests and lint pass, and what should be fixed next.
- Superconductor now automatically sets up the dev environment when you create a project. Any language, any framework. Docker included. If your dev env doesn't need secrets, we auto-apply it to your project, so coding agents can immediately run tests and show live app previews
- Added breadcrumb navigation across the app. With so many ticket implementations going at once, it's nice to immediately know where you are as you're jumping around between them.
- Superconductor repostedGPT 5.5 is the 👑 NEW KING 👑 on our "personal SWE-Bench", based on gold standard PRs into our Ruby on Rails codebase. GLM 5.1 and Kimi K2.6 now beat all Anthropic models (but are slower). And Opus 4.7 is a real head-scratcher! Build your own and see: superconductor.com/blog/agent-ben…
- Recent Activity sidebar now shows GitHub PR status -- open, merged, or closed -- with color-coded badges that update in real time. It's nice to see a sea of purple merged PRs 😎
- You can now define your own branch naming pattern using variables like GitHub username, ticket number, and title. Configure it in your workspace settings to match your team's conventions.
- We added a cheeky little project picker for Slack channels that don’t have a default project (which you can set in Workspace Settings).
- Replying to @SuperconductorBTW, you can set up a "personal SWE-Bench" on your own codebase: 1. Pick several "gold standard" pull requests 2. Select the coding agents to benchmark 3. Coding agents try to re-implement the PRs, and their results are judged by all three frontier LLMs.
