Native Anthropic endpoints, tool-call compatibility, and context-window sizing for local Claude Code.
Last tested: April 2026. See Changelog at the bottom.
If this saves you setup time, follow @renezander030 — practical local AI / coding-agent infrastructure notes, weekly.
Full repo with scripts: github.com/renezander030/local-ai-coding-stack
Production AI Automation Notes #1: Agent Approval Gates — JSON schema, human review, deterministic dispatch, audit logs
Updated 2026-04-28 — JSON schemas, human review, deterministic dispatch, and audit logs for AI agents that touch real systems.
The most common shape of an "AI agent" demo is: model decides → model calls API → side effect happens. That works in a sandbox. It does not work the moment the agent is allowed to send an email to a customer, update a CRM record, or trigger an n8n workflow that hits production.
The pattern that does work is older than agents and is borrowed from financial systems: draft → validate → approve → dispatch → audit. Five steps, five contracts, no exceptions.
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.
April 2026 TLDR setup for Ollama + Gemma 4 on a Mac mini (Apple Silicon) — auto-start, preload, and keep-alive
| #!/usr/bin/env bash | |
| set -euo pipefail | |
| # patch-claude-code.sh — Rebalance Claude Code prompts to fix corner-cutting behavior | |
| # | |
| # What this does: | |
| # Patches the npm-installed @anthropic-ai/claude-code cli.js to rebalance | |
| # system prompt instructions that cause the model to cut corners, simplify | |
| # excessively, and defer complicated work. | |
| # |
| #!/usr/bin/env bash | |
| # | |
| # bootstrap.sh - macOS 輕量終端開發環境一鍵安裝腳本 | |
| # 包含：Ghostty + zellij + yazi + lazygit + fastfetch + Neovim/LazyVim + 現代工具 + oh-my-zsh | |
| # 語言環境：Python(uv/pipx)、Go、Rust(rustup)、Java(Liberica 20 Full+JavaFX)、Node(pnpm+bun)、Ruby | |
| # | |
| # 用法：curl -fsSL https://gist.githubusercontent.com/fuzzbuster/a117689823ba87c27437a44129253dcb/raw/1b2e930ad152a7ceb9fbb7df581b304e2d1685a0/macos_bootstrap.sh | bash | |
| # 或 bash bootstrap.sh | |
| # | |
| # 可重複執行，不會重複安裝或破壞既有配置 |
This document captures learnings about how Agent Skills handle references/, recipes/, and services/ folders, including token budget implications and best practices based on the AgentSkills.io specification and GitHub Copilot implementation behavior.
Agent Skills use a three-tier loading model to efficiently manage LLM context windows:
A complete guide to set up a modern, powerful terminal environment for developers on macOS.
