GRepoWatch
Daily at 07:00 UTC+8Subscribe中文
← All topics

AI paper research and deep-research tools

Research involves more than generating a summary. Decide whether you need an ongoing experimental workspace or a multi-source report around a specific question. Both workflows require source checks; generated conclusions are not validated research findings.

New projects join this topic as the archive updates. Browse the latest recorded projects first.

Choose by your task

7 projects

Last recorded ·

scientific-agent-skills

K-Dense-AI/scientific-agent-skills

Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 250,000+ scientists worldwide. 177 ready-to-use validated skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.

Explore scientific-agent-skills →

Last recorded ·

OpenResearch

alphaXiv/OpenResearch

Manage papers, hypotheses, experiments and artifacts in an ongoing local workspace.

Requirements
Choose the desktop release for your system and configure your coding agent and actual compute environment.
Limitations
Local storage does not mean fully offline operation. Model providers, literature services and remote compute have separate requirements.
Explore OpenResearch →

Last recorded ·

MathModelAgent

jihe520/MathModelAgent

See the project overview and official repository for its purpose and setup.

Explore MathModelAgent →

Last recorded ·

Hyperresearch

jordan-gibbs/hyperresearch

Search multiple sources around a defined question, preserve evidence and produce research reports.

Requirements
Python 3.11–3.14, Claude Code or Codex; check the selected release and project-level installation scope.
Limitations
Spot-check citations. Costs vary with task scope, and advanced long-form modes have experimental conditions.
Explore Hyperresearch →

Last recorded ·

open-science

aipoch/open-science

Open Science by AIPOCH is an open-source, local-first, model-agnostic AI research workbench for macOS, Windows, and Linux, with scientific agents, Python/R notebooks, data connectors, and reproducible provenance.

Explore open-science →

Last recorded ·

academic-research-skills

Imbad0202/academic-research-skills

Academic Research Skills for Claude Code: research → write → review → revise → finalize

Explore academic-research-skills →

Last recorded ·

openhuman

tinyhumansai/openhuman

Your Personal AI super intelligence. A brain that builds a local-first memory of your life, a fantastic orchestrator of agent fleets and workflows, and a deep researcher.

Explore openhuman →

Questions before choosing

Do I need a full workspace to understand one paper?

Not always. Read the abstract, methods and experiments first, ask targeted questions and verify the answers against the original. A full workflow becomes useful for cross-paper comparisons, repeated experiments or traceable reports.

Do these tools include a free model?

Workflow code and model services are separate. Check provider sign-in, model quota, search APIs and compute costs. Installation alone does not satisfy those requirements.

How should I review a research report?

Check primary sources for key conclusions, distinguish author claims from tool inferences, and look for duplicate reposts, outdated material and inaccessible source text. Keep the question, citations and experiment configuration for review.

Projects are grouped using their saved purpose descriptions. Topic membership is a discovery aid, not an installation or performance test. Check each project’s overview and official documentation for requirements, sources and versions.

Explore other topics →