Workspaces¶

Why Workspaces?¶

Zotero collections are great for permanent library organization, but research often needs temporary, cross-cutting groupings — "all papers for my ICML submission", "papers to discuss at lab meeting", or "references for Chapter 3".

Workspaces fill this gap: lightweight, local-only views that don't modify your Zotero library. Each workspace is a TOML file at ~/.config/zot/workspaces/<name>.toml. No API key needed, no syncing side effects.

Create a Workspace¶

zot workspace new llm-safety --description "LLM alignment papers"

Names must be kebab-case (e.g., llm-safety, protein-folding).

Add Items¶

zot workspace add llm-safety ABC123 DEF456 GHI789

Bulk Import¶

zot workspace import llm-safety --collection "Alignment"
zot workspace import llm-safety --tag "safety"
zot workspace import llm-safety --search "RLHF"

Browse¶

zot workspace list                          # All workspaces
zot workspace show llm-safety               # Items with metadata
zot workspace search "reward" --workspace llm-safety

Export¶

zot workspace export llm-safety                       # Markdown
zot workspace export llm-safety --format json         # JSON
zot workspace export llm-safety --format bibtex       # BibTeX

RAG Search¶

Build an index over workspace papers for natural language querying:

Build Index¶

zot workspace index llm-safety

This indexes metadata + PDF full text using BM25.

Query¶

zot workspace query "reward hacking methods" --workspace llm-safety

Returns ranked text chunks from indexed papers.

Semantic Search (Optional)¶

For hybrid BM25 + vector retrieval, configure an embedding endpoint:

export ZOT_EMBEDDING_URL="https://api.jina.ai/v1/embeddings"
export ZOT_EMBEDDING_KEY="your-jina-key"   # 10M free tokens
zot workspace index llm-safety --force      # Rebuild with embeddings
zot workspace query "reward hacking" --workspace llm-safety --mode hybrid

Manage¶

zot workspace remove llm-safety ABC123      # Remove item
zot workspace delete llm-safety --yes       # Delete workspace