Tessera

Tessera

A free control room for running several AI coding assistants at once, side by side, without losing track of which one is doing what.

🔗 Visit Tessera
📁 Editors, IDEs & Dev Tools🗣️ English📅 August 30, 2026

Description

Developers who use AI coding agents heavily often end up running several at once — one fixing a bug, another writing tests, a third exploring a refactor — each in its own terminal tab, which quickly turns into a mess of windows nobody can track. Tessera organizes all of that into one workspace: each agent session gets its own isolated copy of the code (so they can't interfere with each other) and its own place in a visual board that shows what's being worked on, what's in review, and what's done.

Tessera is a free, open-source local workspace for coordinating multiple parallel AI coding agent sessions — Claude Code, Codex and OpenCode are all supported. Each session runs in its own Git worktree to avoid conflicts between parallel tasks, and the app organizes them with collections, persistent tabs, split-pane layouts and a Kanban board (Chat, Todo, Doing, Review, Done). A "Chat View" turns a live raw terminal session into a readable conversation thread, and there's in-app file editing, diff review, git commits and PR creation, plus mobile pairing to keep an eye on sessions from a phone. It's released under the AGPL-3.0 license, runs entirely locally (data stored in a local `~/.tessera/` folder), requires Node.js 20+, and is built with Electron and Next.js.

💬 Our review

The short version: if you regularly run more than one AI coding agent at a time and are tired of juggling terminal tabs to track them, Tessera is a free, well-thought-out control panel for exactly that — with essentially zero cost to trying it since it's open source.

It occupies the same emerging "agent cockpit" niche as tools like Superagent, but Tessera's specific angle is multi-agent, multi-tool orchestration (Claude Code, Codex and OpenCode together) rather than a single-agent experience, plus the Kanban-style task tracking that makes sense once you're running enough parallel sessions to need project-management-style visibility. The Git worktree isolation per session is a genuinely important detail — it's what stops two agents from corrupting each other's changes to the same repo. Being AGPL-3.0 licensed matters if you're considering embedding or reselling it commercially (it has stronger copyleft requirements than MIT), but for personal or internal team use that's a non-issue. As a young project it's worth checking recent commit activity before betting a daily workflow on it, since tooling in this space moves fast and abandoned agent-cockpit projects aren't unheard of.

💰 Pricing

GratuitOpen source AGPL-3.0
Gratuit $0

📊 Global score

53Average
🌐Availability15/100Faible

1 language · 0 platform

📄Profile90/100Excellent

Profile completeness

🤖 AI-enriched data

💰 Pricing model
🆓 Gratuit

100% gratuit et open source (licence AGPL-3.0)

👥 Target audienceDéveloppeurs faisant tourner plusieurs agents de code IA en parallèle (Claude Code, Codex, OpenCode)
🗣️ Languagesen
🌍 Target countriesInternational
👍

Pros

Gère Claude Code, Codex et OpenCode dans une seule interface

Git worktree isolé par session — pas de conflit entre agents

Tableau Kanban intégré (Chat, Todo, Doing, Review, Done)

100% local, données stockées sur la machine

👎

Cons

Licence AGPL-3.0 — attention en cas de réutilisation commerciale

Projet jeune, activité de développement à surveiller

Nécessite Node.js 20+ installé

❓ Frequently asked questions

What is Tessera in one sentence?
How much does it cost?
Does it prevent agents from conflicting with each other?
Which coding agents does it support?
Can I monitor sessions from my phone?
Is it worth the money compared to alternatives?
Which tool should you pick for your case?