Claude Code Mods
Ray Amjad's Claude Code mods: panes, bands, prompt guards, secret handling and memory tools. One marketplace, each mod installs on its own.
Install
/plugin marketplace add ray-amjad/claude-code-modsInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →claude-code-mods
Mods for Claude Code: panes, bands above the prompt, prompt guards, secret handling and memory tools. Each one is a plugin with a hooks module that runs inside the session, so it can see every event as it happens, change it, and draw in the terminal.
Every mod installs on its own. Pick the ones you want.
Install
This repo is a plugin marketplace. Install any mod with one line:
/plugin install <mod> --marketplace ray-amjad/claude-code-mods
Answer y to add the marketplace, then pick a scope. Or add the marketplace once and install from it:
/plugin marketplace add ray-amjad/claude-code-mods
/plugin install cache-clock@claude-code-mods
The mods
| Mod | What it does |
|---|---|
| cache-clock | A one-line band above the prompt: context used, and a countdown until the prompt cache goes cold. |
| decision-tree | When Claude ends a reply with a question, explains the choice in a side pane and lets you explore any option, or your own, in a real forked session. |
| diagram-pane | A side pane that pins every diagram Claude draws, keeps a live map of the work, and lets you go into boxes and change them. |
| dream-pass | A dreaming pass over Claude's memory folder: merges duplicates, rewrites stale entries, removes dead ones and tidies MEMORY.md, with a backup and undo. |
| jev-skill-suggestion | Scores every skill against your prompt with TypeSafe Jev as you type. Observe, block or steer which skill loads. |
| memory-fork | Keeps auto-memory up to date in the background: after each turn Haiku decides whether memory needs a change, and a tool-less fork of the session drafts it. |
| prompt-bouncer | Bounces vague prompts before Claude sees them, and suggests the project's own term. |
| secret-box | Claude uses a secret without seeing it: a masked box saves keys to .env, Claude gets a placeholder, and committing .env is refused. |
| transcript-redactor | Keeps secrets out of the session transcript: each one becomes a stable placeholder, and the real value goes back in on the way into a tool call. |
| vercel-deploy-status | Pins the Vercel deploy queue of the linked project under the prompt, with each deploy's phase and elapsed time. |
Each mod's README shows it running, explains how it works, and lists its options.
secret-box already includes transcript-redactor. Run one or the other, not both.
Work on a mod
Run a mod from its folder without installing it:
claude --plugin-dir plugins/<mod>
Check it before you push:
claude plugin validate plugins/<mod>
claude plugin test plugins/<mod>
License
MIT. See LICENSE.
Similar plugs
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
Documented system prompts from Anthropic - Claude Fable 5.1, Opus 5.5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro,…
A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES.