wanghao9610Community·15 Agent Skills Included

STAR

Structured research workflow from idea to reproducible experiments

Guides a complete research project from topic brainstorming and literature review through planning, code setup, experiment execution, and result analysis. Keeps every plan, run log, and report in project files instead of chat history, so work resumes across sessions and stays auditable. Includes fifteen coordinated skills covering ideation, planning, coding, environment setup, review, and paper-ready documentation.

39|0

↗

npx skills add wanghao9610/STAR --all -g -y

Available:

Gives the agent behavioral rules for careful coding, file-based research record keeping, project layout, runtime resolution, and verification when executing the STAR workflow skills.

All Skills in This Repository (15)

Pure Emerald Level Indicators

📦 In Repo

wanghao9610

star-idea-storm

Guide researchers through structured ideation to produce defensible research topics.

star-plan-decomposer

Decompose research strategies into executable sub-plans with dependency graphs.

star-proj-adopt

Integrates existing AI research projects into the STAR framework via symlinks and historical logs.

star-code-reviewer

Audits source code for research-convention and plan-implementation compliance with evidence-backed reporting.

star-code-architect

Clone reference implementations and generate codearc.md architecture specs.

star-code-release

Consolidate research code and documentation into a public-ready repository format.

star-env-builder

Build and verify Python runtime environments with conda or venv backends.

star-expt-digest

Aggregate experiment analysis reports and execution logs into dated narrative digests.

star-flow-status

Scan plan files, execution logs, and artifact registries to report project status.

star-metd-summarize

Compile fragmented research plans into structured academic method documents.

star-plan-reviser

Audit research plan nodes against execution logs and disk artifacts.

star-refs-reviewer

Fetch and verify bibliographic records from DBLP, Crossref, and Semantic Scholar.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install STAR?▼

Run `npx skills add wanghao9610/STAR --all -g -y` in your terminal to install all fifteen research workflow skills globally.

What does STAR do for a research project?▼

It covers the full research chain: brainstorming a topic, reviewing related work, writing and decomposing a plan, setting up code and environment, running experiments, and analyzing results. Every step leaves auditable files so the work can be resumed or reviewed later.

Can STAR adopt an existing half-finished project?▼

Yes. The star-proj-adopt skill surveys an existing repository read-only, wires in the STAR layout without moving or renaming anything, and records prior work so the plan tree reflects real progress.

Which AI coding tools work with STAR?▼

STAR ships skill copies for Claude Code, Cursor, Codex, Qwen Code, Kimi Code, and other harnesses, all driven by the same shared instructions. It works in both English and Chinese.

Does STAR run heavy training jobs automatically?▼

No. Skills stop before long training runs, large downloads, or costly API calls, and hand the prepared command back to you. Light validation and code changes run automatically after approval.