Depends. Often we discuss stuff on Discord or Slack and the agent's there so it starts a thread, other times I do that or I kick sth off on my phone.
@steipete
-
Debate Examines Why an Agent Might Restrict an MCP Client
By
–
Come on man, you can't tell me that a agent to implement an MCP would go out of its way to add custom code to restrict the client. That's not even sth common or in the training data.
-
Mixing models loses reasoning context, cache, worsens results
By
–
I wonder how this even works, as soon as you mix models you lose the hidden reasoning context, trash the cache and overall get a worse result?
-
Worktreeinclude customizes files copied by codex for new trees.
By
–
Yup! There’s even a worktreeinclude to customize which files codex should copy when creating new trees.
-
Use orchestrator thread to create sub-threads in macOS app
By
–
Use the macOS app. Create an orchestrator thread and tell it to create sub-threads for various work, watch the magic.
-
Kimi, GLM, MiniMax recent releases are good
By
–
Kimi, GLM, MiniMax all had recent releases and are good.
-
GitHub Action triggers codex to compare against VISION.MD and implement
By
–
It's not! It trigger a GitHub Action, spins up codex, compares it against VISION.MD and if a fit, implements it.
-
Agent automatically evaluates and implements open source issues
By
–
If you make an issue on one of our open source projects, an agent will spin up and decide if it's a good idea and will implement it for you.
-
Codex does by default what Claude needs extra work for
By
–
huh, that needs extra work for Claude? Codex just does that by default.
-
Open platform vs lock-in with company-controlled prompts
By
–
depends if you care about an open platform and model choice or being locked in to one company that decides which prompts are okay and which will be blocked or routed to weaker models.