Research workflow skills for the rusen-brain Obsidian vault: paper triage and notes (with ML enrichment), citation traversal, literature surveys, idea exploration, vault primitives (vaultdb, obsidian-cli), and the /think command for Socratic learning sessions.
Explore the citation graph around a seed paper — forward (who cites it), backward (what it cites), or both. Use when the user is building out related work, looking for prior art, or hunting for extensions. Produces grouped, scored suggestions, cross-referenced against the user's Zotero library and rusen-brain vault. The direction parameter has no default — it is always confirmed with the user.
Explore an unfamiliar codebase in read-only mode and produce a concise onboarding report. Use when the user asks to understand a new repo, onboard onto a project, survey what a codebase does, or starts work in an unfamiliar directory. Covers purpose, tech stack, architecture, current git state, active work, and infrastructure. Does not modify anything.
Explore the literature around a research question or half-formed hypothesis. Use when the user is in brainstorm mode — has an idea but no specific paper — and wants to see what the field says. Searches Zotero, arxiv, and Semantic Scholar; clusters findings as supporting, contradicting, adjacent methods, and tangential; identifies gaps. Does not write to the vault unless the user asks.
Orienting survey of a research topic — discover the current landscape and understand how people are working on it. Use when the user enters a new research area and wants a field scan rather than a deep read of any single paper. Optionally anchored by one or more exemplars (e.g., "like Tencent's HY-World 2.0") which disambiguate the topic without scoping the search. Produces a clustered report organized by method, finding, theoretical result, problem framing, or a mix — with an executive overview.
Synthesize a literature review from a set of papers into a note in the rusen-brain vault, matching the existing style (problem framing → thematic clusters with paper counts → synthesis → gaps). Use when the user wants a Zotero collection, tag, or explicit paper list turned into a coherent review. Writes to /home/rusen/Documents/rusen-brain/3-Notes/.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Personal Claude Code plugin bundling a research workflow for the rusen-brain Obsidian vault.
The plugin is named rusen-research. The repo also acts as a single-plugin marketplace, so installation is a two-line claude plugin marketplace add + claude plugin install.
| Skill | Description |
|---|---|
lit-research | Orienting survey of a new research area — discover the landscape, cluster by method / finding / theory / problem framing, produce an executive overview. First-touch skill when entering a new topic. |
paper-triage | Fast SKIM/READ/DEEP/SKIP verdict for a paper from arxiv, Zotero, DOI, or a PDF path. |
paper-notes | Turn a paper (or a Zotero collection/tag) into Article Notes in the rusen-brain vault. For ML/AI papers, also produces an ML Context enrichment block (benchmarks vs SOTA, code, compute, follow-ups, criticisms). Pass mode=enrich-only to produce just the enrichment block without writing a note. |
lit-review | Synthesize a literature review from a set of papers into a vault note: thematic clusters, comparison, synthesis, gaps. |
citation-trace | Explore the citation graph around a seed paper — forward, backward, or both — with Zotero and vault cross-reference. |
idea-deep-dive | Explore the literature around a research question or hypothesis; cluster findings as supporting / contradicting / adjacent / tangential. |
| Skill | Description |
|---|---|
obsidian-cli | Interact with an open Obsidian vault via the obsidian CLI — read, create, search, manage notes, or develop plugins. |
vaultdb | Query, mutate, and traverse Obsidian vaults using vaultdb — a Rust CLI that treats markdown folders as a database with a citation graph. |
| Skill | Description |
|---|---|
context-init | Explore an unfamiliar codebase in read-only mode and produce a concise onboarding report. |
Slash commands invoked explicitly with /<plugin>:<command>:
| Command | Description |
|---|---|
/rusen-research:think | Enter Thinking Buddy mode — Socratic, no first-draft generation, explain-back required, predict-before-reveal. A friction-on learning mode designed to use AI to sharpen the user rather than to ship faster output. Optional argument: a topic to seed the session (e.g. /rusen-research:think B-trees vs B+ trees). Exit by saying ship mode or just answer. |
Skills in the research workflow group follow strict conventions:
/vaultdb and /obsidian-cli. The research skills orchestrate; they don't reinvent vault machinery.[[wiki-links]], not bullet dumps. Vault frontmatter respects the 3-tag schema (type/X + topic/Y + source/Z).paper-triage → paper-notes; idea-deep-dive → citation-trace → lit-review. Skills name each other in their output "recommended next moves."rusen-agent-skills/
├── .claude-plugin/
│ ├── plugin.json # plugin manifest
│ └── marketplace.json # one-plugin marketplace pointing at "./"
├── skills/
│ ├── citation-trace/SKILL.md
│ ├── context-init/SKILL.md
│ ├── idea-deep-dive/SKILL.md
│ ├── lit-research/SKILL.md
│ ├── lit-review/SKILL.md
│ ├── obsidian-cli/SKILL.md
│ ├── paper-notes/SKILL.md
│ ├── paper-triage/SKILL.md
│ └── vaultdb/SKILL.md
├── commands/
│ └── think.md # /rusen-research:think — Thinking Buddy mode
├── README.md
└── LICENSE
# 1. Add this repo as a marketplace (one time)
claude plugin marketplace add https://github.com/rusenbb/rusen-agent-skills
# 2. Install the plugin
claude plugin install rusen-research@rusen-research
# 3. Restart Claude Code
Skills will appear under the rusen-research: namespace (e.g. rusen-research:paper-triage).
To install from a local clone instead:
git clone https://github.com/rusenbb/rusen-agent-skills ~/code/rusen-agent-skills
claude plugin marketplace add ~/code/rusen-agent-skills
claude plugin install rusen-research@rusen-research
Changes to skill files take effect in the current Claude Code session without restart.
npx claudepluginhub rusen-archive/rusen-agent-skills --plugin rusen-researchA growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.