This is a redacted copy of an internal document used at Coefficient Giving. The original document was authored by Chris Webster.
This document is shared in the hope it'll be a useful example for other groups in the community. Coefficient Giving takes no liability for any of the advice here. Note that many tradeoffs we've made (e.g. on security posture, approved apps) may pencil out different for your team.
This copy was made in early January 2026. It will not be updated, so if you're reading this more than a few weeks from now, parts of it may be out of date!
Please share your feedback! What do you do differently, and why? What new things should we try? Mistakes we're making? Please comment on this document or email Chris.
<aside> <img src="/icons/computer_lightgray.svg" alt="/icons/computer_lightgray.svg" width="40px" />
Claude Code is Anthropic's command-line AI tool that runs locally on your computer. Unlike the web interface, it can directly read and modify files on your machine, run terminal commands, and interact with tools like Git and GitHub. Think of it as Claude with direct access to your filesystem.
</aside>
Claude Code requires a paid plan from Anthropic. Check your organization's access policies or sign up at claude.ai.
Desktop App (easier, less powerful): Download from claude.ai/download. Good if you're not comfortable with command-line tools.
Terminal CLI (recommended): Follow the Claude Code quickstart guide.
<aside> ⚠️
Claude Code runs with your user permissions. It can read, modify, and delete any file you can access. This is powerful and potentially dangerous.
</aside>