Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Searches across your Notion workspace, synthesizes findings from multiple pages, and creates comprehensive research documentation saved as new Notion pages. Turns scattered information into structured reports with proper citations and actionable insights.
.claude/skills/bilal140202-notion-research-documentation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 95% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 47% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 16% | 0% |
Enables comprehensive research workflows: search for information across your Notion workspace, fetch and analyze relevant pages, synthesize findings, and create well-structured documentation.
When asked to research and document a topic:
Notion:notion-search to find pagesNotion:notion-fetch to read full page contentNotion:notion-create-pages to write documentationUse Notion:notion-search with the research topic
Filter by teamspace if scope is known
Review search results to identify most relevant pagesUse Notion:notion-fetch for each relevant page URL
Collect content from all relevant sources
Note key findings, quotes, and data pointsAnalyze the collected information:
Use the appropriate documentation template (see reference/format-selection-guide.md) to structure output:
Choose the appropriate format based on request:
Research Summary: See reference/research-summary-format.md Comprehensive Report: See reference/comprehensive-report-format.md Quick Brief: See reference/quick-brief-format.md
By default, create research documents as standalone pages. If the user specifies:
page_id parentdata_source_idSearch filtering: See reference/advanced-search.md Citation styles: See reference/citations.md
"No results found": Try broader search terms or different teamspaces "Too many results": Add filters or search within specific pages "Can't access page": User may lack permissions, ask them to verify access
See examples/ for complete workflow demonstrations:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→fail | 10,247 | 4,635 | -55% | 1 | 1 | 0% | 1,367 | 957 | -30% | 0 | 0 | — |
case-01 | fail→fail | 7,427 | 4,625 | -38% | 1 | 1 | 0% | 1,127 | 972 | -14% | 0 | 0 | — |
case-02 | fail→fail | 19,894 | 4,916 | -75% | 1 | 1 | 0% | 3,090 | 1,008 | -67% | 0 | 0 | — |
case-04 | pass→pass | 9,304 | 8,612 | -7% | 1 | 1 | 0% | 1,413 | 2,080 | +47% | 0 | 0 | — |
case-05 | pass→pass | 11,274 | 8,559 | -24% | 1 | 1 | 0% | 1,963 | 2,270 | +16% | 0 | 0 | — |
case-06 | pass→pass | 10,530 | 6,347 | -40% | 1 | 1 | 0% | 1,494 | 1,723 | +15% | 0 | 0 | — |
case-07 | fail→fail | 7,139 | 7,615 | +7% | 1 | 1 | 0% | 299 | 1,192 | +299% | 0 | 0 | — |
case-08 | fail→pass | 6,084 | 7,220 | +19% | 1 | 1 | 0% | 977 | 1,901 | +95% | 0 | 0 | — |
case-09 | pass→pass | 11,425 | 5,721 | -50% | 1 | 1 | 0% | 1,747 | 1,732 | -1% | 0 | 0 | — |
case-10 | pass→pass | 9,829 | 8,720 | -11% | 1 | 1 | 0% | 1,656 | 2,109 | +27% | 0 | 0 | — |
case-11 | pass→pass | 9,170 | 5,689 | -38% | 1 | 1 | 0% | 1,390 | 1,572 | +13% | 0 | 0 | — |
case-12 | pass→pass | 16,490 | 9,657 | -41% | 1 | 1 | 0% | 2,575 | 2,489 | -3% | 0 | 0 | — |
case-13 | fail→pass | 9,694 | 4,590 | -53% | 1 | 1 | 0% | 1,424 | 1,437 | +1% | 0 | 0 | — |
case-14 | pass→pass | 14,307 | 5,067 | -65% | 1 | 1 | 0% | 2,158 | 1,555 | -28% | 0 | 0 | — |
case-15 | fail→pass | 4,490 | 2,332 | -48% | 1 | 1 | 0% | 615 | 1,099 | +79% | 0 | 0 | — |
case-16 | fail→fail | 13,030 | 7,533 | -42% | 1 | 1 | 0% | 1,815 | 1,871 | +3% | 0 | 0 | — |
case-17 | fail→fail | 6,538 | 1,774 | -73% | 1 | 1 | 0% | 964 | 1,050 | +9% | 0 | 0 | — |
case-18 | fail→fail | 5,947 | 6,046 | +2% | 1 | 1 | 0% | 404 | 1,085 | +169% | 0 | 0 | — |
case-19 | fail→fail | 5,449 | 5,244 | -4% | 1 | 1 | 0% | 315 | 1,046 | +232% | 0 | 0 | — |
case-20 | fail→fail | 6,048 | 5,411 | -11% | 1 | 1 | 0% | 248 | 1,084 | +337% | 0 | 0 | — |
case-21 | pass→pass | 9,018 | 6,587 | -27% | 1 | 1 | 0% | 1,341 | 1,701 | +27% | 0 | 0 | — |
case-22 | pass→pass | 9,984 | 5,414 | -46% | 1 | 1 | 0% | 1,471 | 1,546 | +5% | 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, and 15 counted toward the lift figure. The other 7 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +14 percentage points is the difference between those two pass rates over the 15 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.