Harness Guides
Hands-on guides for operating an AI agent harness: hooks, plugins, permission controls, and the workflows that keep an agent doing real work inside the lines you set.
New to the term? Start with the harness comparison or the Claude Code guide.
Claude Code Hooks: The Complete Guide to Building Hooks in the Desktop App, Including Function Hooks
A start-to-finish guide to Claude Code hooks: where they live, how to add one in the desktop app, exit codes and JSON output, seven copy-paste hooks, and the new function hooks with a working secret redactor.
What /plugin-authoring Unlocks in Claude Code: Plugins From a Sentence
The built-in plugin-authoring skill appears once the function hooks preview is on. What it knows, what it can build (UI rows, spoken summaries, custom tools, guards, cached fetches), prompts to try, and how to iterate.
Install a Secret Redactor Function Hook in Claude Code Desktop With One Prompt
Keep pasted API keys out of the Claude Code transcript while the session can still use them. Turn on the function hooks preview, hand one prompt to the built-in plugin-authoring skill, then test it with a throwaway key.