Xiangyue-Zhang/auto-deep-researcher-24x7

Auto Experiment launches an autonomous THINK→EXECUTE→REFLECT loop that runs deep learning experiments 24/7. Use it to iterate on GPU training, monitor runs at zero LLM cost, and automatically decide next steps from results.

Conf Search finds papers from top AI/ML conferences using Semantic Scholar, letting researchers filter by venue and query, sort by relevance and citations, and analyze results.

Daily Papers searches arXiv for recent papers relevant to a user's research interests and deduplicates recommendations. Use it inside Claude Code or Codex to get a daily ranked briefing of new papers with detailed analysis.

Experiment Status checks the current state of autonomous experiment loops, including progress, recent decisions, training process status, and GPU utilization. Use it to monitor long-running experiments and review results without interrupting execution.

GPU Monitor checks GPU status, running experiments, and available resources by running nvidia-smi and summarizing the output. Use it to quickly see which GPUs are free or in use for training.

Obsidian Sync refreshes an Obsidian dashboard and daily notes from the current state of a Deep Researcher project. Use it when you want progress documentation updated from experiment state, memory logs, and cycle counters.

Paper Analysis performs deep analysis of a single academic paper, extracting figures from the arXiv source and producing a structured review covering problem, method, experiments, and insights.

Progress Report generates structured research progress reports from project memory logs and recent experiment logs. Use it to summarize current status, key insights, next steps, and blockers in autonomous research projects.