Immortalqx

Immortalqx / research-survey-loop

Public

A Codex skill for long-running, multi-round research surveys in Robotics and Embodied AI.

14
0
89% credibility
Found Apr 26, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A workflow skill for AI assistants to conduct persistent, multi-round literature surveys on robotics and embodied AI topics by fetching papers, extracting text, and maintaining organized notes across sessions.

How It Works

1
📚 Discover the Research Helper

You find a handy tool that lets your AI assistant dive deep into topics like robots, AI vision, and navigation by surveying papers over time.

2
🛠️ Add to Your AI Workspace

You place this helper into your AI assistant's special folder so it's ready to use.

3
🔍 Start Your Survey

You chat with your AI: 'Survey recent work on world models for robot navigation, using my notes and PDFs here. Begin round 1.'

4
📥 AI Finds and Gathers Papers

Your AI searches trusted sources, downloads relevant papers, and sorts them into your project folder.

5
📖 AI Reads and Builds Notes

It reads paper sections bit by bit, takes smart notes, updates logs, and grows your survey summary round by round.

6
🔄 Continue Sessions Later

You return anytime and say 'Continue round 2 using the current notes' to keep building without starting over.

🎉 Get Your Full Survey

You now have a complete, organized research overview with all papers, logs, and a polished summary ready to use.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 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 research-survey-loop?

This Python-based Codex skill automates long-running, multi-round literature surveys for robotics, embodied AI, computer vision, world models, navigation, and manipulation. Drop it into your Codex skills directory, prompt with a topic and local PDFs/notes, and it spins up persistent tasks that search arXiv, Semantic Scholar, and targeted publishers via Exa, download papers, chunk PDFs for reading, and iteratively build a survey markdown. You get round-by-round progress in organized workspace folders with task logs, current tasks, and evolving summaries—no manual paper hunting.

Why is it gaining traction?

It stands out as a codex skills library entry with tight Codex GitHub integration for seamless prompting, like "continue round 2," preserving state across sessions unlike one-shot searches. Developers dig the CLI tools for init tasks, fetching sources with deduping, and PDF extraction, plus bilingual docs and hooks for local files. In a sea of basic codex github actions or plugins, this handles multi-round workflows for deep research without babysitting.

Who should use this?

Robotics researchers or embodied AI PhDs surveying recent papers on world models or manipulation. Academics with local notes/PDFs needing structured lit reviews. Teams prototyping codex skills examples for custom research loops in constrained domains.

Verdict

Worth a spin for niche AI lit reviews if you're in Codex already—solid docs and user-facing scripts make it plug-and-play despite 11 stars and 0.9% credibility score signaling early maturity. Test on a small topic first; lacks broad polish but delivers for robotics workflows.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.