MODS FOR CLAUDE CODE
See every move Claude makes.
/plugin marketplace add partalemre/MODZ /plugin install modz@modz
Then add any mod by name: /plugin install savvy-progress@modz. See the mods.
# every mod, in one go
claude plugin marketplace add partalemre/MODZ
for m in modz savvy-progress savvy-flow filetree \
plan-steps handoff transceiver blast-radius \
secret-guard checkpoint test-badge replay-theater \
session-budget cache-tax hud-pro effort-router \
done-ping reflect-mod skins terminal-browser; do
claude plugin install "$m@modz"
done
Start a new session afterwards. The panel opens on the right.
Install MODZ for me. Follow https://partalemre.github.io/MODZ/llms.txt and ask me which mods I want.
Paste it into Claude Code or any coding agent. Every step it runs is safe to repeat.
Claude Code 2.1.295 or laterDesktop app and terminalMIT, free
The film · 54 seconds
One panel. Every mod, live.
It opens beside the chat. One honest line per mod, one click to its pane.
Every agent gets a face.
Progress, cost and time for each one, as it happens.
Risky commands wait for your OK.
You see what would be lost before anything is.
Transceiver · Beta
Your team, inside Claude.
Chat with teammates. Let your agents ask theirs. Zero tokens.
The mods
Install the panel, then any of these. Click a name to copy its install command. Most use no model tokens at all.
Work
- The panel: one live line per mod, one click to each pane.
- The Agents pane and progress band: each subagent's progress, cost and time.
- Orchestrated delivery: the session model plans, tiered worker agents build.
- Files Claude touched this turn, what it is doing now, git status.
- The plan's steps as Claude works through them.
- A handoff note before the context fills, handed to your next fresh chat.
- Team chat, and zero-token agent questions between teammates. Beta.
Safety
- Holds destructive shell commands and shows what they would change.
- Stops keys and tokens in writes, edits, git add and commits.
- A silent git snapshot before each turn's first write. Undo a turn.
- Runs tests and lint after edits and shows the result.
- Replays the last turn's edits step by step, and earlier turns and past sessions.
Meter
- A spend budget per session, with a stop at 100%.
- Keeps the prompt cache warm through breaks and prices a cold send.
- One line: the 5-hour and 7-day limits, context and cost.
- Each turn's effort and model; light prompts can go at low effort.
Comfort
- A notification when a long turn ends or Claude waits on you.
- Saves corrections worth keeping to your CLAUDE.md.
- Themes for the transcript, switched live with /skin.
- A browser inside the terminal that the agent can drive.
Make it yours
- Arrange the panel. Customize moves rows up and hides the ones you do not use. Your layout is kept.
- Get back from anywhere. Every pane has a
← MODZbutton that brings the panel forward. - Add your own mod. Any mod that reports a line gets a row, with no change to MODZ. How.
Installing with an AI agent
Ask your agent to "install MODZ". It can follow llms.txt, which holds these steps:
- Check that
claude --versionis 2.1.295 or later. If not, ask the person to update. - Run
claude plugin marketplace add partalemre/MODZ. - Run
claude plugin install modz@modz, thenclaude plugin install <name>@modzfor each mod the person wants. - Merge optional settings into the
envblock of~/.claude/settings.json, keeping what is there. - Check that
claude plugin listshows each mod enabled, then ask the person to start a new session.
{ "env": {
"MODZ_LANG": "tr", // optional: Turkish interface
"CLAUDE_CODE_ARTIFACT": "1" // transceiver inside an SDK host or IDE
} }
19 mods for Claude Code. One panel.
/plugin marketplace add partalemre/MODZ