Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Sync Zotero references and annotations to Notion databases
.claude/skills/brycewang-stanford-notero-zotero-notion-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-15 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 76% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 51% | 0% |
| case-07 | ✓→✓ | = Same ✓ | 44% | 0% |
Notero is a Zotero plugin that syncs your reference library to a Notion database. When you add or update items in Zotero, Notero automatically creates or updates corresponding Notion pages with metadata (title, authors, DOI, tags), annotations, and notes. Bridges the gap between reference management and knowledge management workflows.
bash# 1. Download .xpi from GitHub releases # 2. In Zotero: Tools → Add-ons → Install Add-on From File # Prerequisites: # - Zotero 7 # - Notion account with a database for papers # - Notion integration token
markdown### Required Properties Create a Notion database with these properties: | Property | Type | Description | |----------|------|-------------| | Title | Title | Paper title (auto-filled) | | Authors | Rich text | Author names | | Year | Number | Publication year | | DOI | URL | Digital Object Identifier | | URL | URL | Paper link | | Abstract | Rich text | Paper abstract | | Tags | Multi-select | Zotero tags | | Item Type | Select | Article, book, etc. | | Zotero URI | URL | Link back to Zotero item | | Date Added | Date | When added to Zotero | | Publication | Rich text | Journal/conference name |
# In Zotero: Tools → Notero Preferences
# 1. Notion Integration Token:
# - Go to notion.so/my-integrations
# - Create new integration
# - Copy the token
# - Share your database with the integration
# 2. Database ID:
# - Open your Notion database
# - Copy the ID from the URL
# - Format: notion.so/<workspace>/<database_id>?v=...
# 3. Property Mapping:
# - Map Zotero fields to Notion properties
# - Configure which fields to sync
# 4. Sync Settings:
# - Auto-sync on add/modify
# - Sync annotations (highlights + notes)
# - Sync attachments (link to PDF)markdown### Automatic Sync 1. Add paper to Zotero (via browser connector, DOI, etc.) 2. Notero detects new item 3. Creates Notion page with metadata 4. Updates on subsequent edits ### Manual Sync 1. Select items in Zotero 2. Right-click → "Sync to Notion" 3. Bulk sync supported ### Annotation Sync 1. Annotate PDF in Zotero reader 2. Highlights and notes sync to Notion page 3. Page references and colors preserved
json{ "notion": { "propertyMapping": { "title": "Title", "creators": "Authors", "date": "Year", "DOI": "DOI", "url": "URL", "abstractNote": "Abstract", "tags": "Tags", "itemType": "Item Type", "publicationTitle": "Publication" }, "syncOptions": { "syncOnModify": true, "syncAnnotations": true, "syncNotes": true, "createBacklink": true, "pageIcon": "📄" }, "formatting": { "authorsFormat": "Last, F.", "maxAuthors": 5, "tagPrefix": "" } } }
markdown### Paper Review Template When a paper syncs, Notero can use a Notion template: ## Summary <!-- Auto-filled abstract --> ## Key Contributions - [ ] Contribution 1 - [ ] Contribution 2 ## Methodology <!-- Your notes --> ## Results <!-- Your notes --> ## Relevance to My Research <!-- Your notes --> ## Questions / Follow-ups - [ ] Question 1 ## Rating: ⭐⭐⭐⭐☆
markdown### Recommended Workflow Zotero (collect) → Notero → Notion (organize + notes) ↓ Notion AI (summarize) ↓ Writing tool (cite) ### Complementary Plugins - **Zotero Better BibTeX** — Citation keys in Notion - **Zotero Style** — Visual organization in Zotero - **Zotero Sci-Hub** — PDF retrieval before annotation - **Notion Web Clipper** — Save web pages alongside papers
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 47,488 | 45,294 | -5% | 1 | 1 | 0% | 3,267 | 4,041 | +24% | 0 | 0 | — |
case-02 | fail→fail | 44,906 | 12,318 | -73% | 1 | 1 | 0% | 2,881 | 3,547 | +23% | 0 | 0 | — |
case-03 | fail→fail | 17,952 | 49,882 | +178% | 1 | 1 | 0% | 3,054 | 4,106 | +34% | 0 | 0 | — |
case-04 | fail→fail | 13,845 | 17,242 | +25% | 1 | 1 | 0% | 2,212 | 3,598 | +63% | 0 | 0 | — |
case-05 | pass→pass | 8,728 | 9,615 | +10% | 1 | 1 | 0% | 1,439 | 2,529 | +76% | 0 | 0 | — |
case-06 | pass→pass | 9,163 | 34,118 | +272% | 1 | 1 | 0% | 1,204 | 1,823 | +51% | 0 | 0 | — |
case-07 | pass→pass | 12,262 | 10,787 | -12% | 1 | 1 | 0% | 1,980 | 2,855 | +44% | 0 | 0 | — |
case-08 | pass→pass | 14,205 | 14,433 | +2% | 1 | 1 | 0% | 2,146 | 3,581 | +67% | 0 | 0 | — |
case-09 | pass→pass | 11,161 | 4,239 | -62% | 1 | 1 | 0% | 1,610 | 1,762 | +9% | 0 | 0 | — |
case-10 | pass→pass | 9,567 | 36,597 | +283% | 1 | 1 | 0% | 1,556 | 2,354 | +51% | 0 | 0 | — |
case-11 | pass→pass | 16,340 | 9,933 | -39% | 1 | 1 | 0% | 1,587 | 2,930 | +85% | 0 | 0 | — |
case-12 | pass→pass | 11,878 | 4,625 | -61% | 1 | 1 | 0% | 1,925 | 1,773 | -8% | 0 | 0 | — |
case-13 | pass→pass | 13,125 | 12,136 | -8% | 1 | 1 | 0% | 1,864 | 3,211 | +72% | 0 | 0 | — |
case-14 | pass→pass | 8,595 | 4,289 | -50% | 1 | 1 | 0% | 1,512 | 1,931 | +28% | 0 | 0 | — |
case-15 | fail→pass | 7,619 | 3,414 | -55% | 1 | 1 | 0% | 1,354 | 1,645 | +21% | 0 | 0 | — |
case-16 | fail→pass | 14,112 | 6,976 | -51% | 1 | 1 | 0% | 2,688 | 2,307 | -14% | 0 | 0 | — |
case-17 | pass→pass | 8,344 | 4,301 | -48% | 1 | 1 | 0% | 1,502 | 1,964 | +31% | 0 | 0 | — |
case-18 | pass→pass | 18,655 | 19,395 | +4% | 1 | 1 | 0% | 2,918 | 3,982 | +36% | 0 | 0 | — |
case-19 | pass→pass | 16,865 | 16,374 | -3% | 1 | 1 | 0% | 2,704 | 3,868 | +43% | 0 | 0 | — |
case-20 | pass→pass | 10,960 | 13,265 | +21% | 1 | 1 | 0% | 2,088 | 3,756 | +80% | 0 | 0 | — |
case-21 | pass→pass | 9,762 | 7,701 | -21% | 1 | 1 | 0% | 1,936 | 2,680 | +38% | 0 | 0 | — |
case-22 | pass→pass | 10,970 | 8,826 | -20% | 1 | 1 | 0% | 1,939 | 2,863 | +48% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +9 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.