Page 2 of 3
Every installed skill spends context just being listed, before you invoke anything. Sort the list by cost, then set the ones you never use to name-only.
Use /init to generate a starter CLAUDE.md, then delete anything inaccurate or obvious before committing.
Cycle into plan mode (Shift+Tab) so Claude explores and proposes an approach for approval before editing.
Instead of guessing which commands to allowlist or giving up and skipping permissions entirely, let Claude read your past sessions and propose the rules you have already been approving.
Declare a Compact Instructions section in CLAUDE.md so both manual /compact and autocompact preserve the project’s critical decisions by default.
Check in project settings so every teammate gets the same allow/deny rules, hooks, and env defaults without redoing /permissions.
Install the GitHub app and mention @claude in review comments so corrections become durable CLAUDE.md rules for every future session.
Add a deterministic PreToolUse hook that rejects destructive shell patterns before Claude can run them.
Auto-compact waits until your context is nearly full, which is well past the point where answer quality starts sliding. Set the window lower and compact while the session is still sharp.
Create a skill that forces a red→green verification loop so Claude cannot claim done without running the suite.
Run claude remote-control on your machine, then connect from the Claude mobile app to approve builds, poke CI failures, and keep coding with full local tools.
A skill that checks a link or reformats a file does not need your session model. Set model, effort and context: fork in the frontmatter so cheap skills stay cheap.
Use /clear between unrelated tasks so leftover debugging does not bias the next feature.
Add a top-level cache_control ephemeral flag so growing conversation prefixes reuse cached tokens and cut API cost and latency.
Add a Notification hook that plays a short sound when Claude is idle or waiting on permission so you can multitask without staring at the terminal.
Turn off MCP servers unrelated to the current project—their tool schemas still consume context at session start.
"Never add comments" in CLAUDE.md plus "make this readable" in your prompt is a contradiction Claude has to resolve before it writes a line. Find those pairs and pick one.
Enable /voice and hold Space to speak complex requirements into the prompt instead of typing walls of text.
Build slash skills that accept issue numbers, branch names, or formats via $ARGUMENTS / $0 so one command covers many cases.
Run /context to break down token consumers—CLAUDE.md, MCP tools, conversation—and prune the expensive ones.
When a skill hits an edge case, append it to a gotchas section in that skill rather than to CLAUDE.md. The lesson then loads only when the skill runs.
Use bundled /loop and /schedule skills for recurring health checks, log watches, or periodic repo tasks without external cron wiring.
Add the GitHub MCP server so Claude can cross-reference issues, PRs, and repo state without leaving the session.
Define custom subagents with `isolation: worktree` so delegated edits happen on a temporary branch checkout.