By jiahao-shao1
Automate AI/ML research and Claude Code workflows: delegate coding/refactoring to OpenAI Codex CLI for fast execution, audit project contexts/docs for compliance, generate daily summaries from sessions/Git/Notion, track experiments in YAML registry, handoff structured overviews, initialize projects with AGENTS.md, integrate Notion LifeOS/Obsidian vault, fetch/analyze papers/BibTeX, orchestrate multi-pane terminals, manage remote GPU clusters.
npx claudepluginhub jiahao-shao1/sjh-skills --plugin sjh-skillsFetch BibTeX entries from arXiv / Semantic Scholar. Triggers: 'fetch bibtex', 'cite this paper', 'arXiv bibtex', '拉 bibtex', '获取引用', '引用这篇论文'. Not for citation format conversion (APA/MLA) or literature review.
cmux multi-pane / multi-agent terminal orchestration via cmux CLI (Bash cannot split panes or spawn agents). Use for parallel panes, splits, spawning Claude/Codex sub-agents, sending keys (incl. ctrl-c) between panes, reading pane output, sidebar updates, browser/markdown panes. Triggers: 'in parallel', 'split pane', 'spawn agent', 'fan out', 'browser pane', 'sidebar', '分屏', '并行', '开个 pane'. Not for plain bash or generic tmux.
Cross-model second opinion from OpenAI Codex on plans or code. Auto-detects plan vs code review, iterates up to 5 rounds. Use for architecture decisions, non-trivial refactors, critical config changes. Triggers: '/codex-review', 'codex review', 'second opinion', 'cross-check', '让 codex 看看', '交叉审查'. Not for trivial fixes or formatting.
Delegate a coding task (debug, implement feature, refactor) to OpenAI Codex CLI directly via codex exec. Faster than codex:rescue agent because it skips the Node companion runtime. Codex writes the code; Claude verifies. Use when you want Codex to actually do the work, not just review. Triggers: '/codex-task', 'codex task', 'let codex do', 'delegate to codex', 'codex implement', 'codex debug', '让 codex 干', '让 codex 实现', '让 codex 修', '丢给 codex', '外包给 codex'. Not for review or second opinion (use codex-review).
Audit project context management (CLAUDE.md, rules, knowledge) for progressive disclosure compliance. Checks that all knowledge files are reachable from rules, detects stale references, orphaned files, and CLAUDE.md index leakage. Triggers: 'audit context', 'check context', 'context hygiene', 'context audit', '检查 context', '审计上下文', 'knowledge 覆盖检查'. NOT for: code review, rule content correctness, or knowledge content quality.
Produce daily/yesterday work summary (日报) by aggregating Claude Code sessions, Git commits, and Notion task data — git log alone cannot do this. Args: today (default), yesterday, 24h, YYYY-MM-DD. Triggers: 'daily summary', 'what did I do today', 'yesterday summary', '日报', '今天干了什么', '工作总结', '某天做了什么'. Not for weekly reports or summarizing files.
Manage ML experiment lifecycle via structured YAML registry — register, record results, compare runs, track status. Triggers: 'register experiment', 'compare experiments', 'experiment results', 'exp init', 'exp list', '注册实验', '对比实验', '查实验', '哪个实验最好', '记录结果'. Not for training monitoring, checkpoints, or launching jobs.
Generate a structured handoff summary for the next session. Triggered when user says "handoff", "交接", or "写个交接".
初始化新项目的 Claude Code 配置。用户说"初始化项目"、"init project"、"配置 Claude Code"时触发。
User's personal Notion LifeOS (PARA: tasks, notes, projects, Make Time journals) with pre-configured DB IDs + API access — required for any LifeOS interaction. Use for adding tasks, taking notes, journaling, querying todos, searching notes, creating projects, reflection. Triggers: 'add task', 'take a note', 'what do I need to do', 'search my notes', '加个任务', '查笔记', '待办', '记录想法', '反思', '目标偏移'. Not for Notion API docs or workspace admin.
Obsidian Second Brain (dual-zone vault) Claude Code integration. Use for loading project context, capturing thoughts, morning planning, querying linked notes. Triggers: 'obsidian', 'second brain', 'vault', 'capture thought', 'today plan', '第二大脑', '今日计划', '记录想法'. Not for Notion (use notion-lifeos) or papers (use scholar-agent).
Deep critical analysis of one specific paper — dissect experimental setup, extract numbers, evaluate claims against a hypothesis. Triggers: 'analyze this paper', 'deep dive into', '分析这篇论文', '论文深度分析', 'what does this paper really show', '这篇论文说明了什么', arXiv ID with analysis intent. Not for paper discovery (use scholar-agent), quick summaries, or literature surveys.
Self-review for paper paragraphs — diagnose v1 issues and give a revision direction (intro, abstract, method, related work). Triggers: 'review this intro', 'check this paragraph', 'paper paragraph self-review', 'help me look at intro / abstract', 'polish this intro', 'paper self-review', '帮我看 intro', 'paper 段落自检', 'review 这段 paper'. Not for: generating a draft from scratch — write v1 yourself first, then iterate.
Strategic project review — auto-discovers strategy docs (vision, roadmap, decisions, notes), produces five-dimension snapshot. Use for direction review, alignment check, milestone status, pre-meeting overview. Triggers: 'project review', 'strategic review', 'where is the project at', '项目全景', '审视战略', '项目现在什么状态'. Not for code/PR review.
Remote GPU cluster ops via `rca` CLI — run commands, transfer files, GPU/node inspection, mutagen sync. Trigger for any remote execution, training, GPU, cluster status, sync, or tail-log task even when 'cluster' isn't said. Triggers: cluster, GPU, rca, training, node, sync code, tail log, 集群, 节点, 远程执行, 同步代码, 跑训练, 看日志, GPU 占用.
Academic paper discovery / filtering / reading via Scholar Inbox API + notebooklm-py CLI. Use for browsing today's papers, recommendations, rate/collect, deep-read via NotebookLM, ask questions about papers, trending. Triggers: 'scholar inbox', 'paper digest', '看论文', '读论文', '论文推荐', 'rate papers', '今天有什么论文', 'deep read paper', '帮我筛选论文'. Not for PDF reading, lit reviews, or general notes.
Check whether documentation needs updating after code changes. Triggered when user says "sync docs", "文档同步", or "检查文档".
Fetch any URL as clean markdown. Use instead of WebFetch for JS-rendered pages, login-required platforms (Twitter/X, zhihu, reddit, weibo, xiaohongshu, bilibili, etc.), and complex pages. Routes known platforms through OpenCLI (browser login state), others through Jina Reader / defuddle.md / markdown.new. Invoke when the user provides a URL to read, extract, summarize, or convert to markdown.
Share bugs, ideas, or general feedback.
Curate auto-memory, promote learnings to CLAUDE.md and rules, extract proven patterns into reusable skills.
General purpose Claude Code toolkit with research agents, skills, and introspection commands
63 universal AI skills for Claude Code — skill creation, deep research, strategic planning, product management, content processing, UI/UX design, Obsidian knowledge management, and orchestration workflows
Core skills: ecosystem guide, skill creator, research patterns, session reflection, and plugin development. Includes UserPromptSubmit hook for forced skill evaluation.
Create and resume session handoffs for multi-session work continuity.
Skill and agent authoring tools: generate, audit, and improve Claude Code skills, commands, and agents
English | 中文
A collection of Claude Code skills for research workflow automation.
| Skill | Description |
|---|---|
| scholar-agent | Taste-aware research agent — personalized paper discovery via Scholar Inbox + hallucination-free deep reading via NotebookLM |
| cmux | Agent-friendly terminal built on Ghostty — multi-agent orchestration via split panes, spawn sub-Claude-Code instances, built-in browser, markdown preview, sidebar progress reporting |
| daily-summary | Daily work summary — aggregates Claude Code sessions, git commits, and Notion tasks into a timeline-style Chinese report |
| notion-lifeos | Notion life management — PARA method + Make Time journaling, with natural language task/note/journal CRUD via Notion API |
| web-fetcher | Web page → clean markdown with 5-layer fallback: Jina Reader → defuddle.md → markdown.new → OpenCLI (platform-specific with login state) → raw HTML |
| init-project | Initialize Claude Code project config — CLAUDE.md scaffolding, agent templates, and research profile setup |
| project-review | Project strategy panoramic review — auto-discover strategy docs and generate a 5-dimension snapshot (vision, roadmap, blockers, related work, next steps) |
| remote-cluster-agent | Remote GPU cluster operations — edit code locally, run commands remotely via Go daemon + rca CLI (exec / batch / cp / nodes), persistent SSH connection pool, node health monitoring |
| codex-review | Cross-model review — send your plan or code diff to OpenAI Codex for independent verification, iterative Claude↔Codex feedback until approved (max 5 rounds) |
| codex-task | Delegate a coding task (debug / implement / refactor) directly to OpenAI Codex CLI. Single-shot codex exec with gpt-5.5 + xhigh effort and full sandbox bypass by default; Claude verifies the diff after. Faster than codex:rescue (skips the Node companion runtime) |
| paper-analyzer | Deep critical paper analysis — causal chain methodology (现象→实验设置→归因→解法), NotebookLM-grounded reading, optional research framework mapping |
| paper-self-review | Paper paragraph self-review — three-axis check (logic jumps / repetition / premature method detail), v1 → v2 deterministic iteration |
| experiment-registry | ML experiment lifecycle management — structured YAML registry with CLI for registering experiments, recording benchmarks, comparing results, and tracking status |
| handoff | Session handoff summary — prints a structured context summary (status, decisions, pitfalls, next steps) directly in the conversation for seamless session continuity |
| sync-docs | Documentation sync checker — scans recent code changes and reports which docs (knowledge base, registry, CLAUDE.md, rules, README) need updating. Report only, no auto-modify |
| context-audit | Context management hygiene — audits the three-layer architecture (CLAUDE.md / rules / knowledge) for progressive disclosure compliance. Detects orphaned knowledge, stale references, and CLAUDE.md index leakage. Read-only |
| bibtex-fetch | Fetch correct BibTeX entries from arXiv (by ID) or Semantic Scholar (by title search). Batch fetch, custom bibkeys, exponential backoff. Zero dependencies (Python stdlib only) |
| obsidian-brain | ⏸️ On Hold — Obsidian Second Brain with dual-zone vault. Paused in favor of enhancing notion-lifeos with reflection commands |
/plugin marketplace add jiahao-shao1/sjh-skills
/plugin install sjh-skills@sjh-skills
/reload-plugins
Plugin auto-updates from GitHub on startup. No manual sync needed.
Tell Codex:
Fetch and follow instructions from https://raw.githubusercontent.com/jiahao-shao1/sjh-skills/refs/heads/main/.codex/INSTALL.md
Or manually:
git clone https://github.com/jiahao-shao1/sjh-skills.git ~/.codex/sjh-skills
mkdir -p ~/.agents/skills
for skill in ~/.codex/sjh-skills/skills/*/; do
ln -sf "$skill" ~/.agents/skills/$(basename "$skill")
done
Detailed docs: .codex/INSTALL.md
npx skills add jiahao-shao1/sjh-skills --skill scholar-agent
npx skills add jiahao-shao1/sjh-skills --skill cmux
All skills at once:
npx skills add jiahao-shao1/sjh-skills