This is a slow walkthrough of OpenAI tool calling using only raw API calls and PowerShell.
The point:
- No agent framework is required.
- No harness is required.
- The model does not execute your code.
- The model asks for a tool call.
| { | |
| "acts": [ | |
| "ACT.OVERGROWTH", | |
| "ACT.HIVE", | |
| "ACT.GLORY" | |
| ], | |
| "ascension": 1, | |
| "build_id": "v0.103.2", | |
| "game_mode": "standard", | |
| "killed_by_encounter": "NONE.NONE", |
| -- ~/.wezterm.lua | |
| -- Cross-platform WezTerm config (Windows / macOS / Linux). | |
| -- Gist: https://gist.github.com/romgenie/a6acd389dc53c78802b971058cd281f1 | |
| local wezterm = require 'wezterm' | |
| local mux = wezterm.mux | |
| local act = wezterm.action | |
| local config = wezterm.config_builder() | |
| -- ============================================================ |
A pattern for building personal knowledge bases using LLMs.
This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.
Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.
Keep main stable. Treat feature branches as the default path for any meaningful change.
codex/ prefix when it fits the task, such as codex/add-spectator-delay.todo.md updated while the work is in progress.