I have two jobs, and in my personal time I wanna make sure OpenClaw works for ANY model provider, local or cloud.
@steipete
-
E2E Testing Strategy for AI Model Harness Compatibility
By
–
We constantly run e2e tests to ensure harness changes don't break with most common models. That's all my org and my key… well, was.
-
OpenClaw Compatibility Challenges with Anthropic Models Ahead
By
–
Yeah folks, it's gonna be harder in the future to ensure OpenClaw still works with Anthropic models.
-
Chat Models Have Limits Unlike Codex Systems
By
–
Everything has limits. the chat model isn’t agentic, it’s more like codex.
-
Mac mini for local AI models isolation use cases
By
–
yeah that's different, most folks don't buy mac mini for local models tho, just for isolation.
-
Anthropic’s System Prompt Blockers Growing Increasingly Complex
By
–
Anthropic's randoms system prompt blockers are getting weirder and weirder.
-
Security Approvals Framework for AI Systems Iteration
By
–
The default is no approvals, you can set this to yes. We do wanna get security right tho and will iterate on approvals until they are bullet proof. I know it's painful.
-
OpenAI Subscriptions Now Fully Supported in OpenClaw
By
–
Yes, OpenAI subs are fully supported in OpenClaw.
-
Safety Company Alignment Issues Compared to Competitors
By
–
I’m not surprised, defo an alignment issue tho, funny how the safety company here is far worse than the competition.