Introducing a new open-source tool called TeamCopilot: a coding agent for teams (https://t.co/3O1pBYf6Xn).
Think Claude Code, but shared across your entire team and running on your cloud. A few key differences compared to other AI agents:
• Multi-user environment -> everyone uses the same agent setup. Configure once, the whole team can use it.
• Skill & tool permissions -> control who can use which skills and tools through the agent. Example: allow only certain people in the team to use a skill for making server config changes.
• Approval workflow -> anyone can create tools/skills, but engineers in the team must approve them before the agent can even see them.
• Fully auditable -> chat sessions can’t be deleted by users and are stored on your server.
• Use it anywhere -> web UI lets you talk to the agent even when you're away from your work machine.
The aim here is create a safe and user friendly environment for all team members to leverage AI agents!
It works with OpenAI and Claude models using API keys or subscriptions!
This is just a start and a lot more features coming up! 👀
@AodenTeoMT This sounds incredible. I hope the generated voices are actually that expressive and not just cherry picked for the demo video.
Do you have voice changer feature that can change my accent, voice, and remove the “uhmmm” etc from my videos?
These are some of the new features added to https://t.co/MK0cvhwwy0:
- A secret manager for your team so that LLMs never have to see your secrets directly.
- Scheduling cronjobs: Run AI agents with a todo-list on a schedule for automating business processes
- Calling workflows via APIs
- Long running agents: Give agents a long running task that can run for hours until the task is complete. Since teamcopilot runs on your servers, you can check in via your mobile, so you can work on the go.
@simonarabian@felixrieseberg Yea, it’s a pipeline that uses a bunch of stuff: frontier ai models + graph algorithms + cubicasa model + SAM2 model. Each of them individually arent enough, but a combination of them work to give about 70-90% accuracy.
@thenowhereway Cause: react is the most popular frontend framework -> NextJS is the most popular way to use react now -> vercel is the easiest to deploy NextJS apps in.
@enjojoyy Once I gave codex a task to implement a big feature in codebase with 250 tests. It built the feature, then ran the existing tests and most of them failed. So it started to fix them one by one and that took about 3 hours.
@garrytan That’s why I’m working on https://t.co/GxSoNF8yOa. It’s a coding agent for teams. One place to create a knowledge base + useful AI skills and workflows for your team.
@paulg@t_blom But then here domain knowledge will be in people’s chat history with the ai across various tools and platforms. How would you extract that?
@webdevcody I use an ec2 instance in which I have setup my dev env and on top of that run https://t.co/GxSoNF8yOa. It’s a coding agent with a web UI, so I can keep working on the go.
@ethshea@ycombinator@DraftedAI Do consider trying https://t.co/pbaXjG7taQ next time! It helps you visualise the floor plan in first / third person 3d view quickly.
@ycombinator@DraftedAI And if you would like to view the your plan in first person / third person 3d (like walking around in a video game), checkout https://t.co/vAGeFSi8qC
@DraftedAI And for those that want to visualise their floor plans in first or third person 3d view (like in a video game), checkout https://t.co/pbaXjG7taQ