Two experiments in the next @openclaw to address some "GPT is lazy" issues: 1) Strict mode: agents.defaults.embeddedPi.executionContract = "strict-agentic" This tells GPT-5.x to keep working: read more code, call tools, make changes, or return a real blocker instead of
@steipete
-
Experimental GPT modes: strict mode and Codex server testing
By
–
there’s two experimental modes for gpt we are currently testing. You can enable strict mode that nudges the model to keep working; and you can switch to codex app server as harness.
-
Creator Defends Influential Work, Seeks Alternative Solutions
By
–
Open for suggestions how that would work. Or are you blaming me for creating something that inspired millions?
-
State Position Against Powerful Dangerous Technology
By
–
Sure, but should we? My state so far has been no – it is too powerful and therefore dangerous.
-
Liking Tech While Rejecting Its Toxic Culture
By
–
I like the tech, just not some aspects of the culture around it, the grifting, the gambling.
-
Great Conference Speaker Worth Booking Recommendation
By
–
i was just about to leave the conference, and he caught me in the lobby. he's a great speaker btw. you should book him. https://
youtu.be/fdbXNWkpPMY?si
=UWL7Jxctkm_qvXzj
… -
E2E Testing Across AI Models for OpenClaw Framework
By
–
Nah I use codex, but for OpenClaw we run e2e tests with all models to ensure there's no harness regressions.
-
Software Requires Human Judgment Beyond Agent Automation
By
–
Software still needs thinking. Agents automate the manual work, but not the taste, the design, the saying no.