A working Python agent loop with risk-tiered tools, a SQLite approval queue, deny-by-default timeouts and structured logs. Standard library only, tested, and ready to plug into a real LLM.
Building an AI Agent with Human Approval Gates
Related
No model decides whether a model made something up
Cast is my job-search system. An LLM tailors each CV and cover letter to a role, and drafts replies...
"Don't reveal it" is not an access control. RAG permissions belong in retrieval.
The gap between the demo and production In the tutorial version of RAG, every document in...
Writing a pi extension that tells you when the agent is done
I keep switching away while pi works. It's a long edit, or a test suite, and there's nothing to...