WenyuChiouWenyuChiou/research-hub/SKILL.md

research-hub

Skill67 starsChanged 7 months agoMITView on GitHub

Use research-hub to operate Zotero, Obsidian, and NotebookLM research workflows through CLI, MCP, REST, and dashboard. Trigger when the user asks to find papers, ingest sources, organize a literature review, build Obsidian research notes, upload sources to NotebookLM, inspect clusters, generate AI briefs, or maintain a research vault.

In this repository14 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- gap-to-topicSkill
- literature-triage-matrixSkill
- notebooklm-brief-verifierSkill
- paper-memory-builderSkill
- paper-summarizeSkill
- research-context-compressorSkill
- research-design-helperSkill
- research-hub-multi-aiSkill
- research-hubSkill
- research-project-orienterSkill
- research-workflow-orchestratorSkill
- zotero-library-curatorSkill

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
Default language policy: respond in the user's language, but generate durable research notes and metadata in English unless the user asks …
line 10
Very specific:
Run research-hub doctor first when setup or runtime behavior is unclear.
line 64
Very specific:
Use --no-nlm for smoke tests or when NotebookLM browser automation is not configured.
line 65
Very specific:
Prefer import-folder for analysts, internal knowledge bases, local PDFs, DOCX, Markdown, TXT, and URL sources.
line 66
Distinctive:
Prefer Zotero-backed commands for DOI/arXiv-heavy academic literature workflows.
line 67
Distinctive:
Do not delete clusters without previewing cascade impact first.
line 68
Distinctive:
Treat the vault as user-owned local data.
line 69

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

SKILL.md

research-hub

research-hub is an AI-operable research workspace. It can use Zotero, Obsidian, and NotebookLM together, but it is also useful when the user only has two of them.

Default language policy: respond in the user's language, but generate durable research notes and metadata in English unless the user asks otherwise.

Choose The Workflow

User has	Prefer this path
Zotero + Obsidian + NotebookLM	Full `research-hub auto "topic"` workflow
Zotero + Obsidian	Search/add/backfill papers, write Markdown notes, skip NotebookLM with `--no-nlm`
Obsidian + NotebookLM	`import-folder`, dashboard, NotebookLM bundle/upload/generate
Zotero + NotebookLM	Zotero-backed paper selection and NotebookLM operations
No accounts yet	`research-hub dashboard --sample`

First Commands

research-hub dashboard --sample
research-hub setup
research-hub auto "topic" --no-nlm
research-hub serve --dashboard
research-hub doctor --autofix

For local files without Zotero:

research-hub setup --persona analyst
research-hub import-folder ./papers --cluster my-local-review
research-hub serve --dashboard

AI Host Integration

For MCP hosts, configure:

{ "mcpServers": { "research-hub": { "command": "research-hub", "args": ["serve"] } } }

Install host-specific skill files when useful:

research-hub install --platform claude-code
research-hub install --platform cursor
research-hub install --platform codex
research-hub install --platform gemini

For Hermes, OpenClaw, or other hosts with SKILL.md/rules support, copy the relevant skills/<name>/ directories manually or inline the skill text into the host's instructions. Use MCP/REST for tool calls when no installer target exists.

Operating Rules

Run research-hub doctor first when setup or runtime behavior is unclear.
Use --no-nlm for smoke tests or when NotebookLM browser automation is not configured.
Prefer import-folder for analysts, internal knowledge bases, local PDFs, DOCX, Markdown, TXT, and URL sources.
Prefer Zotero-backed commands for DOI/arXiv-heavy academic literature workflows.
Do not delete clusters without previewing cascade impact first.
Treat the vault as user-owned local data; avoid overwriting notes unless the user explicitly asks.

Useful Docs

README.md: public project overview.
docs/first-10-minutes.md: persona-based onboarding.
docs/mcp-tools.md: MCP tool reference.
docs/ai-integrations.md: Claude/Cursor/Codex/Gemini integration paths.
docs/import-folder.md: local file ingest.
docs/notebooklm.md: NotebookLM setup and automation.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free