the whole reason I started building it: I kept hitting my Claude limit halfway through tasks, and the work would just sit there until the reset.
curious what everyone else does when that happens. do you wait it out, switch to Codex or Cursor and re-explain everything, pay for extra usage, or do you have some smarter setup I haven't thought of?
what I ended up building lets the agents I already have hand the task to each other when one runs out. but I'm genuinely curious if there's a workflow I'm missing.
Other tools run agents side by side. Friday makes them work together: when Claude runs out mid-task, another agent you already run picks it up with the progress so far, and agents message each other instead of you copy-pasting between terminals.
Friday Work is a multiplayer AI workspace for teams and coding agents. Share a session and teammates join, watch and steer live. Pick up your existing Claude Code and Codex sessions with the same context, or run them beside Friday's agents, each in its own worktree, and hand work between them. Build saved agents with skills and triggers. Connect 50+ tools. Review the diff, commit, push and open the PR in one place. Mac and Windows, free to start; Pro and Max add frontier models.
Friday WorkOpen-source harness for coding agents that work as a team
