sjh-skills
Research workflow automation for papers, experiments, and clusters
All Skills in This Repository (17)
Pure Emerald Level Indicatorspaper-self-review
Diagnose research paper paragraphs across five axes and direct revisions.
scholar-agent
Automate academic paper discovery, filtering, and deep reading via Scholar Inbox API and NotebookLM CLI.
project-review
Auto-discover strategy documents and generate a five-dimension snapshot.
web-fetcher
Convert web pages into clean Markdown via OpenCLI or Jina Reader.
handoff
Generate a structured handoff summary from conversation, git diff, and task list.
context-audit
Audits CLAUDE.md, rules, and knowledge for progressive disclosure compliance and generates a structured remediation report.
obsidian-brain
Load Obsidian vault context into Claude Code with dual-zone governance.
sync-docs
Audits code changes and flags outdated or missing documentation.
notion-lifeos
Convert natural language prompts into Notion tasks, notes, projects, and Make Time entries.
daily-summary
Aggregate Claude Code sessions, Git commits, and Notion tasks into a timeline report.
paper-analyzer
Generate a structured causal-chain analysis report from an arXiv paper.
experiment-registry
Register and manage ML experiments in a YAML-based registry.
Frequently Asked Questions
FAQPage SchemaHow to install sjh-skills?â–¼
Run `npx skills add jiahao-shao1/sjh-skills --all -g -y` in your terminal to install the full collection globally, or add `--skill <name>` to install a single skill.
How to automate a research paper reading workflow?â–¼
The scholar-agent skill fetches your personalized Scholar Inbox digest, filters papers by your interests, and deep-reads them through NotebookLM with source-grounded answers.
Can it manage ML experiments and GPU clusters?â–¼
Yes. The experiment-registry skill tracks runs and benchmarks in YAML files, and remote-cluster-agent lets you run commands and sync code on remote GPU nodes with low latency.
Does sjh-skills work with Claude Code and Codex?â–¼
Yes. Every skill follows the standard SKILL.md format and works in Claude Code, Codex, Cursor, and other compatible agents.
Can it help with paper writing and citations?â–¼
Yes. paper-self-review checks drafts for logic and framing issues, paper-analyzer does deep critical analysis, and bibtex-fetch pulls accurate citations from arXiv and Semantic Scholar.
Related Repositories in Education & Research
View All in Education & Research→learn-claude-code
Learn to build AI agent harnesses from scratch
ai-agents-for-beginners
Learn to build, deploy, and secure AI agents step by step
33-js-concepts
Master 33 essential JavaScript concepts with tested, verified guides