Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute Linktree secondary workflow: Analytics & Insights. Trigger: "linktree analytics & insights", "secondary linktree workflow".
.claude/skills/jeremylongshore-linktree-core-workflow-b/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 57% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 233% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -17% | 0% |
Pull click tracking data, measure conversion rates, and extract audience insights from your Linktree profile. Use this workflow when you need to understand which links drive traffic, where your audience comes from, and how engagement trends over time. This is the secondary workflow — for link creation and profile management, see linktree-core-workflow-a.
typescriptconst analytics = await client.analytics.get({ profile_id: profile.id, period: 'last_30_days', granularity: 'daily', }); console.log(`Total views: ${analytics.views}`); console.log(`Total clicks: ${analytics.clicks}`); console.log(`CTR: ${(analytics.clicks / analytics.views * 100).toFixed(1)}%`); analytics.daily.forEach(d => console.log(` ${d.date}: ${d.views} views, ${d.clicks} clicks`) );
typescriptconst linkStats = await client.analytics.byLink({ profile_id: profile.id, period: 'last_7_days', sort: 'clicks_desc', }); linkStats.forEach((s, i) => console.log(`#${i + 1} ${s.title}: ${s.clicks} clicks (${s.unique_visitors} unique, CTR ${s.ctr}%)`) );
typescriptconst audience = await client.analytics.audience({ profile_id: profile.id, period: 'last_30_days', }); console.log('Top locations:', audience.locations.slice(0, 5).map(l => `${l.country}: ${l.pct}%`)); console.log('Top referrers:', audience.referrers.slice(0, 5).map(r => `${r.source}: ${r.visits}`)); console.log('Devices:', `Mobile ${audience.devices.mobile}% / Desktop ${audience.devices.desktop}%`);
typescriptconst report = await client.analytics.report({ profile_id: profile.id, period: 'last_30_days', metrics: ['views', 'clicks', 'ctr', 'unique_visitors', 'top_referrers'], format: 'json', }); console.log(`Report period: ${report.start_date} to ${report.end_date}`); console.log(`Highest CTR link: ${report.top_link.title} (${report.top_link.ctr}%)`); console.log(`Total unique visitors: ${report.unique_visitors}`);
| Issue | Cause | Fix | |-------|-------|-----| | 401 Unauthorized | Expired or invalid bearer token | Re-authenticate via OAuth flow | | 403 Analytics unavailable | Free-tier profile without analytics | Upgrade to Pro plan for analytics access | | 404 Profile not found | Wrong profile_id | Verify with client.profiles.list() | | 422 Invalid period | Unsupported date range format | Use last_7_days, last_30_days, or ISO dates | | Empty daily array | Profile has zero traffic in range | Widen the date range or verify profile is published |
A successful workflow produces daily view/click time series, a ranked list of link performance with CTR, audience demographics broken down by location, referrer, and device type, and a summary conversion report identifying top-performing links.
See linktree-sdk-patterns for OAuth setup and rate-limit handling.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | fail→pass | 15,937 | 7,597 | -52% | 1 | 1 | 0% | 1,890 | 1,349 | -29% | 0 | 0 | — |
case-01 | fail→pass | 16,067 | 13,947 | -13% | 1 | 1 | 0% | 1,909 | 2,914 | +53% | 0 | 0 | — |
case-02 | fail→pass | 14,524 | 13,371 | -8% | 1 | 1 | 0% | 1,752 | 2,742 | +57% | 0 | 0 | — |
case-03 | fail→pass | 21,526 | 16,393 | -24% | 1 | 1 | 0% | 1,032 | 3,434 | +233% | 0 | 0 | — |
case-04 | pass→pass | 13,225 | 11,652 | -12% | 1 | 1 | 0% | 1,273 | 1,800 | +41% | 0 | 0 | — |
case-05 | pass→pass | 13,112 | 10,166 | -22% | 1 | 1 | 0% | 1,313 | 1,741 | +33% | 0 | 0 | — |
case-06 | pass→pass | 10,484 | 12,945 | +23% | 1 | 1 | 0% | 924 | 2,018 | +118% | 0 | 0 | — |
case-07 | fail→pass | 15,259 | 8,070 | -47% | 1 | 1 | 0% | 1,812 | 1,496 | -17% | 0 | 0 | — |
case-08 | pass→pass | 13,460 | 9,194 | -32% | 1 | 1 | 0% | 1,577 | 1,702 | +8% | 0 | 0 | — |
case-10 | fail→pass | 22,159 | 9,161 | -59% | 1 | 1 | 0% | 2,304 | 1,708 | -26% | 0 | 0 | — |
case-11 | fail→pass | 16,349 | 9,398 | -43% | 1 | 1 | 0% | 1,679 | 1,541 | -8% | 0 | 0 | — |
case-12 | fail→pass | 14,189 | 8,260 | -42% | 1 | 1 | 0% | 1,847 | 1,536 | -17% | 0 | 0 | — |
case-13 | fail→pass | 10,795 | 2,349 | -78% | 1 | 1 | 0% | 952 | 1,293 | +36% | 0 | 0 | — |
case-14 | pass→pass | 18,588 | 4,530 | -76% | 1 | 1 | 0% | 2,166 | 1,469 | -32% | 0 | 0 | — |
case-15 | pass→pass | 11,898 | 5,402 | -55% | 1 | 1 | 0% | 1,598 | 1,822 | +14% | 0 | 0 | — |
case-16 | pass→pass | 9,163 | 2,812 | -69% | 1 | 1 | 0% | 1,573 | 1,241 | -21% | 0 | 0 | — |
case-17 | fail→pass | 15,690 | 7,984 | -49% | 1 | 1 | 0% | 1,609 | 1,451 | -10% | 0 | 0 | — |
case-22 | fail→pass | 17,280 | 7,653 | -56% | 1 | 1 | 0% | 1,690 | 1,353 | -20% | 0 | 0 | — |
case-18 | pass→pass | 9,755 | 4,196 | -57% | 1 | 1 | 0% | 1,563 | 1,460 | -7% | 0 | 0 | — |
case-19 | fail→pass | 9,228 | 3,558 | -61% | 1 | 1 | 0% | 1,694 | 1,367 | -19% | 0 | 0 | — |
case-20 | fail→pass | 11,847 | 9,666 | -18% | 1 | 1 | 0% | 1,986 | 1,515 | -24% | 0 | 0 | — |
case-21 | fail→pass | 11,766 | 4,759 | -60% | 1 | 1 | 0% | 2,244 | 1,593 | -29% | 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 21 counted toward the lift figure. The other 1 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 +64 percentage points is the difference between those two pass rates over the 21 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.