Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Notion expert for product management workflows. Use for database design for PRDs/OKRs/Roadmaps/Decisions, property and view design, page architecture, Notion REST API, and sync with Jira, Linear, and GitHub.
.claude/skills/borghei-notion-pm/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 76% | 0% |
Master-level expertise in using Notion as the documentation and operational backbone for a product team: PRDs, OKRs, roadmaps, decision logs, sprint reviews, 1:1 notes, customer research, and sync patterns with Jira, Linear, and GitHub. Covers Notion's database model, view design, formula and rollup patterns, page architecture, and REST API.
Notion's PM value comes from its database model: every meaningful PM artifact (a PRD, an OKR, a roadmap row, a decision) becomes a row in a typed database, with relations to other databases. This unlocks linked views (one source of truth, many surface presentations), rollups (aggregate child progress into a parent), and a queryable API. The job of a Notion PM expert is to design these databases up front, restrain ad-hoc page creation, and tie the workspace into Jira/Linear/GitHub so the artifacts stay current without manual maintenance.
Pull the reference that matches the task; keep this file lean and load detail on demand.
In Scope: Notion workspace and Teamspace setup, database design for PRDs/OKRs/Roadmap/Decisions/Reviews/1:1s, view design, property and relation modeling, page architecture, Notion REST API authoring, sync pattern design with Jira/Linear/GitHub, governance and review cadences.
Out of Scope: Notion administration (SCIM, SAML, billing) at the workspace level beyond Teamspace setup. Jira space and project configuration (hand off to jira-expert/ and confluence-expert/). Linear configuration (hand off to linear-expert/). Strategic OKR setting (hand off to execution/brainstorm-okrs/). Story splitting and backlog refinement (hand off to execution/backlog-refinement/, execution/story-splitting/).
Limitations: Notion's API rate limits (~3 req/s) make it unsuitable for very high-throughput sync; batch and queue. Status properties cannot have options created via the API in lower plan tiers; pre-create options manually. Rollups cannot reference rollups in older API versions; chain via formulas where needed. Page-level permissions can be overridden in unexpected ways by Teamspace-level changes. Notion is not a real-time collaboration substitute for chat; do not try to replace Slack with comments.
| Integration | Direction | What Flows | |---|---|---| | jira-expert/ | Notion ↔ Jira | PRD-to-Epic creation; Jira issue embeds in PRD pages | | confluence-expert/ | Migration | When moving from Confluence to Notion, structure and content mapping | | linear-expert/ | Notion ↔ Linear | PRD-to-Project creation; Linear progress rollups into Notion | | execution/create-prd/ | PRD → Notion | PRD scaffolder output rendered as Notion blocks via --format notion | | execution/brainstorm-okrs/ | OKR → Notion | OKR validator output written to the OKR DB | | execution/outcome-roadmap/ | Roadmap → Notion | Roadmap rows materialized in the Roadmap DB | | execution/daci-framework/ | Decision → Notion | DACI decisions logged in the Decisions DB | | execution/status-update-generator/ | Notion → Status | Weekly status pulls aggregates from PRDs/OKRs DBs | | execution/release-notes/ | Notion → Release | Shipped PRDs feed release notes input | | senior-pm/ | Notion → Portfolio | OKR and Roadmap DBs feed portfolio reports | | discovery/interview-synthesis/ | Research → Notion | Customer research database populated from interview synth output |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→pass | 13,249 | 14,848 | +12% | 1 | 1 | 0% | 2,241 | 3,692 | +65% | 0 | 0 | — |
case-05 | pass→pass | 16,883 | 18,195 | +8% | 1 | 1 | 0% | 2,941 | 4,618 | +57% | 0 | 0 | — |
case-01 | fail→fail | 15,454 | 17,446 | +13% | 1 | 1 | 0% | 2,781 | 4,509 | +62% | 0 | 0 | — |
case-02 | fail→pass | 13,006 | 8,256 | -37% | 1 | 1 | 0% | 2,117 | 2,691 | +27% | 0 | 0 | — |
case-03 | fail→fail | 12,740 | 23,680 | +86% | 1 | 1 | 0% | 2,124 | 5,444 | +156% | 0 | 0 | — |
case-06 | fail→fail | 15,323 | 14,402 | -6% | 1 | 1 | 0% | 2,458 | 3,936 | +60% | 0 | 0 | — |
case-07 | pass→pass | 10,937 | 11,331 | +4% | 1 | 1 | 0% | 1,980 | 3,346 | +69% | 0 | 0 | — |
case-08 | fail→pass | 15,270 | 16,684 | +9% | 1 | 1 | 0% | 2,718 | 4,631 | +70% | 0 | 0 | — |
case-09 | pass→pass | 14,242 | 30,150 | +112% | 1 | 1 | 0% | 2,234 | 4,006 | +79% | 0 | 0 | — |
case-10 | pass→pass | 17,097 | 17,154 | +0% | 1 | 1 | 0% | 2,543 | 4,126 | +62% | 0 | 0 | — |
case-11 | fail→pass | 14,151 | 11,714 | -17% | 1 | 1 | 0% | 2,200 | 3,420 | +55% | 0 | 0 | — |
case-12 | pass→pass | 14,542 | 14,404 | -1% | 1 | 1 | 0% | 2,240 | 3,607 | +61% | 0 | 0 | — |
case-13 | pass→pass | 10,648 | 8,384 | -21% | 1 | 1 | 0% | 1,911 | 3,051 | +60% | 0 | 0 | — |
case-14 | pass→pass | 16,417 | 14,725 | -10% | 1 | 1 | 0% | 2,502 | 3,667 | +47% | 0 | 0 | — |
case-15 | pass→pass | 13,518 | 11,989 | -11% | 1 | 1 | 0% | 2,055 | 3,340 | +63% | 0 | 0 | — |
case-16 | pass→pass | 15,550 | 14,545 | -6% | 1 | 1 | 0% | 2,271 | 3,660 | +61% | 0 | 0 | — |
case-17 | pass→pass | 11,243 | 11,096 | -1% | 1 | 1 | 0% | 2,169 | 3,397 | +57% | 0 | 0 | — |
case-18 | pass→pass | 15,113 | 18,145 | +20% | 1 | 1 | 0% | 2,436 | 4,487 | +84% | 0 | 0 | — |
case-19 | pass→pass | 9,487 | 8,557 | -10% | 1 | 1 | 0% | 1,748 | 3,252 | +86% | 0 | 0 | — |
case-20 | pass→pass | 15,935 | 14,230 | -11% | 1 | 1 | 0% | 2,335 | 3,825 | +64% | 0 | 0 | — |
case-21 | fail→pass | 12,893 | 13,507 | +5% | 1 | 1 | 0% | 2,096 | 3,695 | +76% | 0 | 0 | — |
case-22 | pass→pass | 11,534 | 10,390 | -10% | 1 | 1 | 0% | 1,854 | 3,289 | +77% | 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 +23 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.