Guides / AI agents
AI agent guides.
How to give Claude Code, Codex, Gemini, and Copilot a memory that survives the session, written as notes you can read.
Gemini CLI persistent memory: making context survive between sessions
Gemini CLI persistent memory explained: how GEMINI.md context files load, where /memory add writes, and how to keep agent memory readable and portable.
2026-08-30
How to set up AI agent config on a new laptop
How to set up AI agent config on a new laptop: what to move, what to leave behind, and how to restore Claude Code settings without losing your rules.
2026-08-30
GitHub Copilot CLI memory: how to make context survive between sessions
GitHub Copilot CLI memory does not survive the session. Here is what persists between sessions, and how to give the agent a store it reads and writes.
2026-08-30
Codex CLI persistent memory: how to make Codex remember between sessions
Codex CLI persistent memory explained: what AGENTS.md keeps, what each new session drops, and how to store durable facts in markdown files you own.
2026-08-30
How to see what your AI agent remembers
How to see what your AI agent remembers: find the instruction files it loads, read every stored decision, and retire the memories that went stale.
2026-08-30
Why does my AI agent forget everything between sessions?
Why does my AI agent forget everything between sessions? What the context window drops, what your tools keep on disk, and how to fix it with plain files.
2026-08-30
How to give Claude Code a persistent memory
Claude Code carries some state between sessions, but not a decision record. Here is how to build a Claude Code persistent memory you can read and edit.
2026-08-30
AI agent memory for code review: remembering feedback and conventions
Build AI agent memory for code review so your coding agent remembers past feedback, team review conventions, and the suggestions you already rejected.
2026-08-30
How to give AI agents persistent memory
How to give AI agents persistent memory using plain markdown files the agent writes and reads back, so decisions and facts survive between sessions.
2026-08-30
MCP memory server vs plain markdown files for agent memory
MCP memory server vs plain files: how each stores what your AI agent remembers, where each one fails, and how to choose without locking your notes away.
2026-08-30
How to set up shared memory across AI coding agents
Shared memory across AI coding agents: keep one markdown store that Claude Code, Codex, Gemini CLI, and Copilot all read, write, and stay in sync with.
2026-08-30
How to sync Claude skills across computers
Claude Code skills are folders of markdown on disk. Four reliable ways to sync Claude skills across computers, with the exact commands for each.
2026-08-30
How to sync CLAUDE.md between machines
Three ways to sync CLAUDE.md between machines: a git dotfiles repo, a file-sync service, or a synced notes vault. Real commands and the tradeoffs.
2026-08-30
How to sync AGENTS.md across computers
Three practical ways to sync AGENTS.md across computers, what git already handles for you, and how to stop two machines from quietly overwriting each other.
2026-08-30
How to share agent instructions with your team
Two working ways to share agent instructions with your team: a committed AGENTS.md for repo rules, and one synced shared folder for everything else.
2026-08-30
Vector database vs markdown for agent memory
A practical comparison of vector database vs markdown for agent memory, covering exact recall, auditing, embeddings, and when each one earns its place.
2026-08-30