One free gateway to hundreds of AI providers. When one hits its limit, it routes you to the next model.
// you asked for this
The five plugins I run in Claude Code so I stop watching my usage limits: a free model router, memory across sessions, fewer tokens per answer, Anthropic's own setup scan, and a skill that learns from your corrections. Every link and the setup are below.
Open it, and see what is inside.
Three steps you paste into Claude Code.
The usual problems, and the fix for each.
// get it
Every link, the install for each one, how to check it works and what to try first. All five are free to install. None of them is mine; the guide is.
Open The Setup GuidePDF . opens in a new tabOne free gateway to hundreds of AI providers. When one hits its limit, it routes you to the next model.
Memory across every Claude Code session, so you stop re-explaining your project every morning.
Compresses logs, files and tool output before they reach the model. Same answer, fewer tokens.
Anthropic's official plugin. Scans your codebase and recommends the hooks, skills and MCP servers you are missing.
Watches how you work, logs every correction and proposes updates to your skills. You approve each change.
// start today
Paste these into Claude Code (the app or the terminal), not a plain chat window: it needs to run commands on your computer.
Step 1 of 3 . start here
Claude Code reads each README and installs them one by one. It stops and asks before anything that needs an account or a key.
Install these 5 Claude Code plugins for me, one at a time. For each one: read its README first, use only the install method it gives, and stop and ask me before anything that needs an account, an API key or a paid plan. 1. https://github.com/cheaper-inference/OmniRoute 2. https://github.com/thedotmack/claude-mem 3. https://github.com/headroomlabs-ai/headroom 4. https://github.com/anthropics/claude-plugins-official (the claude-code-setup plugin) 5. https://github.com/rebelytics/one-skill-to-rule-them-all (the task-observer skill) When you are done, show me how to check each one works.
Step 2 of 3
Open Claude Code in your project and paste this. Claude Code Setup answers with the hooks, skills and MCP servers worth adding. It only reads; it changes nothing.
Recommend automations for this project.
Step 3 of 3
Installing it is not enough. This turns it on for every session, so it can start logging your corrections.
Turn on task-observer for every session: add the activation instruction from its references/environments.md to my CLAUDE.md, then tell me how I will know it is running.
// get unstuck
All five are open source and free to install. OmniRoute only routes to providers you connect, and free tiers have their own limits. Claude-Mem's installer offers a 30-day trial of its hosted memory; you can skip it and use your own key or plan instead. The guide shows both.
No. It logs what it sees and proposes updates. You review them, and nothing goes live until you install the update.
These are for Claude Code, which runs commands on your computer. A plain chat window cannot install them.
They are popular, open-source projects, but they are still other people's code. Read each README, install only from the links on this page, and keep API keys out of your chats. The guide has a safety page.
OmniRoute (cheaper-inference), Claude-Mem (thedotmack), Headroom (headroomlabs-ai), Claude Code Setup (Anthropic) and Task Observer (rebelytics) are made by their owners and are not affiliated with Avi Grondin. Claude and Claude Code are trademarks of Anthropic and do not endorse this page.
// what next . paid . optional

You came to get more out of Claude Code. The Playbook shows how to point AI at the rest of your business: ads, sales and delivery.
I wrote The AI Scale Playbook for the rest: ads, SEO, your website, cold email, DMs, AI calling, video, sales and finance.
All sales are final. You get instant access to digital files, so there are no refunds.
See what is in the Playbook$27 . one payment . next pageNot for you? Close this page. Everything above stays free.