SeeleAI

SeeleAI / Thoth

Public

🐦 Thoth — Dashboard-First Orchestration Runtime for Autoresearch

agent autoresearch claude-code codex dashboard
22
2
100% credibility
Found Apr 27, 2026 at 22 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Thoth provides a dashboard-focused system to orchestrate AI-assisted research with persistent tasks, contracts, and verifiable outcomes across chat tools.

How It Works

1
🔍 Discover Thoth

You find Thoth in your AI chat tool's marketplace and add it to help manage research projects.

2
📁 Pick your project folder

Tell Thoth to set up your folder, and it creates a special workspace just for your ideas.

3
🖥️ Your dashboard appears

A beautiful screen opens showing tasks, progress bars, and everything in one place.

4
✅ Choose a ready task

Pick from clear tasks that are waiting, like research steps or reviews.

5
▶️ Start the task safely

Click to run it, and watch it happen live with updates right on your screen.

🎉 See results and decide next

Everything finishes with proof it worked, ready for your review or more steps.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 22 to 22 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 Thoth?

Thoth, named after the ancient Egyptian god of wisdom and writing (thoth god, thoth gott, thoth meaning), is a Python runtime for dashboard-first orchestration in autoresearch. It locks drifting AI agent work into durable runs, strict task contracts, and reviewable verdicts via a local web dashboard showing live progress, Gantt charts, and activity logs. Install as a Claude Code or Codex plugin, init a repo with /thoth:init, then run tasks like /thoth:run --task-id task-1 and monitor via /thoth:dashboard.

Why is it gaining traction?

Its dashboard-first design surfaces parallel runs, stale states, and mechanical acceptances that chat UIs hide, fixing persistence rot in autoresearch orchestration. Strict --task-id commands enforce contracts across hosts, with read surfaces like status, report, and doctor for quick audits. Python scripts and FastAPI backend deliver instant visibility without vendor lock-in.

Who should use this?

Research engineers running agent-driven experiments, teams using AI for repo changes needing audit trails, or devs switching between Claude Code and Codex seeking workflow parity. Suited for autoresearch projects where thoth symbol of knowledge tracking matters more than free-form chats.

Verdict

Promising niche tool at v0.1.4 with crisp docs and MIT license, but 22 stars and 1.0% credibility signal early maturity—test in a side repo first. Strong for Python autoresearch if you need runtime ledgers over hype.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.