gavraz

gavraz / recon

Public

tmux-native dashboard for managing Claude Code agents

37
1
100% credibility
Found Mar 15, 2026 at 37 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Recon provides a terminal dashboard for monitoring, switching, creating, and managing multiple AI coding agent sessions with both a table view and a fun pixel-art Tamagotchi-style visual display.

How It Works

1
🔍 Discover recon

You hear about recon, a handy terminal dashboard that makes juggling multiple AI coding helpers a breeze without leaving your workspace.

2
⚙️ Set it up

You quickly add recon to your tools, ready to manage your AI sessions right from the terminal.

3
📋 Open the session list

Launch recon to see a simple table listing all your AI helpers, their projects, statuses like working or waiting, and when they were last active.

4
🐾 Switch to creature view

Press 'v' for a delightful pixel-art dashboard where each AI agent appears as a cute animated creature – green blobs bustling, orange ones waving for input!

5
➡️ Jump between helpers

Use simple keys to navigate, zoom into groups, switch to any agent needing attention, or end one that's done.

6
✨ Start new or resume old

Easily create a fresh AI helper in your current project or pick up a past session to continue right where you left off.

🎉 Master your AI team

Now you effortlessly oversee and direct multiple AI coding buddies from one spot, staying productive without tab-switching chaos.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 37 to 37 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is recon?

Recon is a Rust-built tmux-native dashboard for overseeing multiple Claude Code agents right in your terminal. It scans tmux panes and Claude's JSON logs to show live status—working, waiting for input, idle, or new—plus git branches, token usage, and models, letting you switch, kill, spawn, or resume sessions with single keys. No more alt-tabbing; glance at a table view or pixel-art Tamagotchi creatures grouped by repo for instant agent reconnaissance.

Why is it gaining traction?

Unlike basic tmux wrappers, recon delivers animated visuals on a side monitor, git-aware grouping, and smart status detection from Claude's TUI bars, making multi-agent workflows feel alive. CLI commands like `recon next` jump to input-blocked agents, `recon resume` picks from history, and JSON output enables scripting—perfect for tmux keybinds that pop dashboards over any session. Devs dig the zero-friction monitoring for Claude agents without leaving the flow.

Who should use this?

Tmux-heavy backend devs juggling Claude sessions across repos for refactoring, debugging pipelines, or code reviews. Frontend teams running parallel agents on feature branches, or anyone using Claude Code as a recon github tool for rapid prototyping—especially if you're tired of hunting PIDs or context windows manually.

Verdict

Grab it if you're deep in tmux and Claude; the docs and tmux integration shine, but at 37 stars and 1.0% credibility, it's early—test in a sandbox before production swarms. Solid MIT start for agent orchestration.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.