Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Zotero workflow automation with custom actions and tags
.claude/skills/brycewang-stanford-zotero-actions-tags-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -1% | 0% |
Zotero Actions & Tags is a powerful Zotero plugin that enables workflow automation through event-triggered actions. Define custom rules that execute when items are added, modified, opened, or tagged — such as auto-tagging by collection, running scripts on PDF open, auto-renaming attachments, or sending notifications. Transforms Zotero from a passive reference manager into an active research workflow engine.
bash# Download .xpi from GitHub releases # In Zotero 7: Tools → Add-ons → Install Add-on From File
markdown### Event → Condition → Action **Events** (triggers): - Item added to library - Item added to collection - Item modified - Tag added/removed - Attachment opened - Item selected **Conditions** (filters): - Item type (article, book, etc.) - Collection membership - Tag presence/absence - Field value match (regex) - Date range **Actions** (responses): - Add/remove tags - Set field values - Run JavaScript - Open URLs - Copy to clipboard - Show notifications
json{ "name": "Auto-tag ML papers", "event": "add_to_collection", "condition": { "collection": "Machine Learning" }, "actions": [ {"type": "add_tag", "tag": "#ml"}, {"type": "add_tag", "tag": "#to-read"} ] }
json{ "name": "Track reading", "event": "open_attachment", "condition": { "attachment_type": "application/pdf" }, "actions": [ {"type": "remove_tag", "tag": "#to-read"}, {"type": "add_tag", "tag": "#reading"}, {"type": "set_field", "field": "extra", "value": "LastOpened: ${date}"} ] }
json{ "name": "Flag high-impact", "event": "item_modified", "condition": { "field": "extra", "regex": "Citations:\\s*(\\d{3,})" }, "actions": [ {"type": "add_tag", "tag": "#high-impact"}, {"type": "add_tag", "tag": "#priority"} ] }
javascript// Run custom JavaScript on trigger // Example: Auto-format author names { "name": "Format authors", "event": "item_added", "action": { "type": "script", "code": ` const creators = item.getCreators(); // Log to Zotero debug console Zotero.debug('New item by: ' + creators.map(c => c.lastName).join(', ')); ` } }
json[ { "name": "New → To-Read", "event": "item_added", "actions": [{"type": "add_tag", "tag": "#to-read"}] }, { "name": "To-Read → Reading", "event": "open_attachment", "condition": {"has_tag": "#to-read"}, "actions": [ {"type": "remove_tag", "tag": "#to-read"}, {"type": "add_tag", "tag": "#reading"} ] }, { "name": "Reading → Done", "event": "tag_added", "condition": {"tag": "#done"}, "actions": [ {"type": "remove_tag", "tag": "#reading"} ] } ]
json[ { "name": "Tag by journal", "event": "item_added", "condition": { "field": "publicationTitle", "regex": "Nature|Science|Cell" }, "actions": [ {"type": "add_tag", "tag": "#top-journal"} ] }, { "name": "Flag recent papers", "event": "item_added", "condition": { "field": "date", "after": "2024-01-01" }, "actions": [ {"type": "add_tag", "tag": "#recent"} ] } ]
markdown### Settings (Edit → Preferences → Actions & Tags) - **Enable/disable** individual actions - **Action priority** — execution order when multiple match - **Logging** — debug action execution - **Import/Export** — share action sets with collaborators - **Keyboard shortcuts** — trigger actions manually
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 14,383 | 6,750 | -53% | 1 | 1 | 0% | 2,643 | 2,878 | +9% | 0 | 0 | — |
case-23 | pass→pass | 11,160 | 15,822 | +42% | 1 | 1 | 0% | 2,188 | 3,887 | +78% | 0 | 0 | — |
case-01 | fail→pass | 14,102 | 5,896 | -58% | 1 | 1 | 0% | 2,720 | 2,341 | -14% | 0 | 0 | — |
case-02 | fail→pass | 14,905 | 6,408 | -57% | 1 | 1 | 0% | 2,165 | 2,704 | +25% | 0 | 0 | — |
case-04 | fail→pass | 12,131 | 7,293 | -40% | 1 | 1 | 0% | 2,137 | 2,490 | +17% | 0 | 0 | — |
case-05 | pass→pass | 23,490 | 8,583 | -63% | 1 | 1 | 0% | 3,639 | 2,734 | -25% | 0 | 0 | — |
case-06 | fail→pass | 16,934 | 5,755 | -66% | 1 | 1 | 0% | 2,802 | 2,775 | -1% | 0 | 0 | — |
case-07 | fail→pass | 14,753 | 4,375 | -70% | 1 | 1 | 0% | 3,254 | 2,269 | -30% | 0 | 0 | — |
case-08 | fail→pass | 8,120 | 4,538 | -44% | 1 | 1 | 0% | 1,175 | 2,179 | +85% | 0 | 0 | — |
case-09 | fail→pass | 7,834 | 2,008 | -74% | 1 | 1 | 0% | 1,310 | 1,739 | +33% | 0 | 0 | — |
case-10 | fail→pass | 12,420 | 7,795 | -37% | 1 | 1 | 0% | 2,292 | 2,573 | +12% | 0 | 0 | — |
case-11 | pass→pass | 12,078 | 4,037 | -67% | 1 | 1 | 0% | 2,234 | 2,134 | -4% | 0 | 0 | — |
case-12 | pass→pass | 11,808 | 4,312 | -63% | 1 | 1 | 0% | 1,933 | 2,093 | +8% | 0 | 0 | — |
case-13 | pass→pass | 11,700 | 6,138 | -48% | 1 | 1 | 0% | 1,558 | 2,303 | +48% | 0 | 0 | — |
case-14 | fail→pass | 12,818 | 2,979 | -77% | 1 | 1 | 0% | 1,938 | 1,813 | -6% | 0 | 0 | — |
case-15 | fail→pass | 10,696 | 5,502 | -49% | 1 | 1 | 0% | 1,834 | 2,170 | +18% | 0 | 0 | — |
case-16 | pass→pass | 11,160 | 4,416 | -60% | 1 | 1 | 0% | 1,682 | 2,106 | +25% | 0 | 0 | — |
case-17 | pass→pass | 16,290 | 13,234 | -19% | 1 | 1 | 0% | 2,512 | 3,537 | +41% | 0 | 0 | — |
case-18 | fail→pass | 14,500 | 4,094 | -72% | 1 | 1 | 0% | 2,522 | 2,144 | -15% | 0 | 0 | — |
case-19 | fail→pass | 11,480 | 7,589 | -34% | 1 | 1 | 0% | 2,136 | 2,527 | +18% | 0 | 0 | — |
case-20 | pass→pass | 9,516 | 5,465 | -43% | 1 | 1 | 0% | 1,403 | 2,326 | +66% | 0 | 0 | — |
case-21 | pass→pass | 15,715 | 18,286 | +16% | 1 | 1 | 0% | 2,541 | 4,334 | +71% | 0 | 0 | — |
case-22 | pass→pass | 12,027 | 11,507 | -4% | 1 | 1 | 0% | 2,414 | 3,643 | +51% | 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. 23 cases were attempted. The headline lift of +57 percentage points is the difference between those two pass rates over the 23 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.