What is agent-diet?
agent-diet is a Claude Code mod by VedantAndhale that runs search-only subagents (Explore by default) on Haiku instead of your main model; a model Claude asks for still wins, and /agent-diet picks the types and the model. It is listed under Agents & automation and has 10 GitHub stars.
| Type | Claude code mod |
|---|---|
| Author | VedantAndhale |
| Repository | VedantAndhale/claude-pro-kit |
| Popularity | 10 GitHub stars |
| Category | Agents & automation |
| Security | Low risk (100/100) |
| Rank | #511 of 5,272 Claude Code mods; #55 in Agents & automation |
| License | MIT |
| Last commit | 2026-10-09 |
| Stars this week | +4 |
| Hooks | 3 events |
| Sees | subagent spawns |
| Draws on | nothing |
Is agent-diet safe to install?
agent-diet scores 100/100 on the MCPMods security check: Low risk. No checks raised a concern. It passed 8 of 8 checks.
Mod footprints come from claude plugin validate, which lists every event a mod hooks and every API call it makes without running it.
| Check | Result |
|---|---|
| No network access | Pass |
| Starts no processes | Pass |
| Writes no files or config | Pass |
| Doesn't read environment variables | Pass |
| Only sees the events it hooks | Pass |
| Doesn't drive Claude | Pass |
| Passes claude plugin validate | Pass |
| Has an open-source license | Pass |
API calls in its code
| Call | What it means |
|---|---|
$.command.register | |
$.store.get | keeps its own state |
$.store.set | keeps its own state |
$.ui.status | draws interface |
$.ui.toast | draws interface |
How do I install agent-diet?
- Check your Claude Code version. Mods need Claude Code 2.1.287 or later and are on by default.
claude --version
- Read its footprint. This lists every event it hooks and every API call it makes, without running it.
git clone https://github.com/VedantAndhale/claude-pro-kit claude plugin validate ./claude-pro-kit/plugins/agent-diet
- Install it. Run this in your terminal, then /reload-plugins in an open session.
claude plugin marketplace add VedantAndhale/claude-pro-kit && claude plugin install agent-diet@claude-pro-kit
Or ask Claude to set it up
Paste this into Claude Code and it installs agent-diet for you:
Install the "agent-diet" Claude Code mod from https://github.com/VedantAndhale/claude-pro-kit/tree/main/plugins/agent-diet. Run claude plugin validate on it first and show me what it can access before you enable it.
More from VedantAndhale/claude-pro-kit
This repository has at least 9 Claude Code mods on MCPMods.
- answer-pane: Explain, plan and ELI5 pages drawn natively in a side pane: diagrams in place, plan decisions as buttons, Respond fills (Review before installing)
- budget-guard: Holds a prompt back once your 5-hour or weekly usage is at or over a limit you set (90% by default); sending it again go (Low risk)
- cache-clock: Shows how long the prompt cache stays warm, and when it expires, exactly how many tokens your next message will re-send (Low risk)
- cmd-diet: Adds quiet flags to noisy shell commands before they run (git status, pytest, cargo, npm install, mvn, curl, wget, docke (Low risk)
- collision-guard: Asks before Claude edits a file another chat on this machine changed in the last 30 minutes. Each chat keeps its own sma (Review before installing)
- compact-keeper: After a compaction, adds one message with exact facts from before it: your latest prompt in full, up to 4 earlier ones, (Low risk)
- context-xray: A pane with the exact breakdown of what fills Claude's context: system prompt, tools, memory files, skills, messages, an (Low risk)
- gh-account: Runs each git push, pull, fetch, clone and gh command as the logged-in GitHub account that owns or belongs to the reposi (Low risk)