posthog/ai-plugin

choosing-trend-or-slope-view

Clarify how to visualize change over a time range before building a trend.

Zobacz źródło

Jak używać

  1. Skopiuj polecenie instalacji i sprawdź źródła przed instalacją we własnym środowisku agenta.
  2. W oryginale sprawdź wymagane narzędzia, dane dostępowe i obsługiwane środowiska.
  3. Opisz dane wejściowe i oczekiwany rezultat; porównaj wynik ze źródłami.
npx skills add https://github.com/posthog/ai-plugin --skill choosing-trend-or-slope-view

Przed rozpoczęciem

  • Instalacja nie obejmuje subskrypcji modelu, limitu API ani uprawnień do plików.
  • Gwiazdki i instalacje pokazują aktywność, a nie niezależny test jakości lub bezpieczeństwa.
Oryginalny dokument Skill

Treść z repozytorium z zachowaniem nagłówków, przykładów, kodu, tabel, linków i obrazów.

Choosing a trend line vs a slope view

"How did X change between A and B?" is ambiguous. Two charts answer two different questions, so clarify before you build unless the user already named one:

  • Change over time (line) — the value at every interval across the range.

Shows the path: dips, spikes, when it moved. This is the default trend.

  • Start vs end (slope) — only the first and last point, one line per series

connecting them. Shows the net change and, across many series, which rose, which fell, and any rank flips — without the noise of the path between.

When the request could be either, ask a short either/or, e.g.:

Do you want to see how it moved across the whole period (a line chart), or just the change from the start to the end (a slope chart)?

If the user clearly wants one — "just tell me how much it grew start to end" → slope; "show me the trend / when did it spike" → line — skip the question and build it.

How to render each

Both come from the same TrendsQuery over the same date range — the slope is that series collapsed to its first and last point, not a different query.

Change over time → line

Default trends behavior. Create or run a TrendsQuery and leave trendsFilter.display as ActionsLineGraph (the default, "change over time"):

json
{
  "kind": "TrendsQuery",
  "series": [{ "kind": "EventsNode", "event": "$pageview", "math": "total" }],
  "dateRange": { "date_from": "2025-01-01", "date_to": "2025-03-31" },
  "trendsFilter": { "display": "ActionsLineGraph" }
}

Start vs end → slope

Run the trend with posthog:query-trends. The result card Max renders has a Line / Bar / Slope view toggle — switch it to Slope to show each series as a single line from its first to its last point, with the per-series change in the legend. Tell the user they can flip to the Slope view on the result.

The slope view is best for a clean before→after comparison, especially with several series/categories whose relative movement matters. Pick a date range whose two ends are the points you want compared (the slope uses the first and last interval).

Important limits

  • Two surfaces, one computation. Both the inline slope (Max's query-trends

result card) and the saved-insight slope (ChartDisplayType.SlopeGraph) show the same thing: the first interval's value vs the last interval's value, at the chosen group-by interval. The grouping defines the slope - group by month to compare the first vs last month, by day for the first vs last day. Use the inline view for a quick before-and-after on a result you're already looking at; use the saved display to persist it on a dashboard. A still-accumulating final period is shown as-is with a dashed connector, the same affordance the line chart uses for an incomplete tail.

  • For period-over-period on a single series (this month vs last), a line with

compareFilter: { "compare": true } overlays the two periods; a slope is the better fit when comparing the endpoints of many series at once.

Powiązane zadania

z tego samego repozytorium

Więcej Skills

Wszystkie Skills
posthog
Oficjalne

composing-grid-canvases

Compose PostHog grid canvases — widget grids (including the user's home canvas) built from reusable component canvases. Use when a task asks to add, fill, move, resize, or remove a widget on a grid or home canvas, to compose a whole canvas of widgets from one ask, to build a reusable widget/component, or when a placement id or grid canvas id is the target. Covers the component store search → configure → fork → build ladder, the component placement contract (size, configSchema), the placement lif

Instalacje / 1h
1
GitHub Stars
96
Aktualizacja
31 paź
posthog
Oficjalne

consuming-endpoints-from-client-code

Wire a PostHog endpoint into a client app or SDK. Covers fetching the OpenAPI spec, generating a typed client with openapi-generator or @hey-api/openapi-ts, sending the right auth header, shaping the variables payload (HogQL codename vs insight breakdown property), handling rate-limit and materialised-endpoint error responses. Use when the user says "how do I call my endpoint", "generate a client for this", or "what auth header do I use".

Instalacje / 1h
1
GitHub Stars
96
Aktualizacja
31 paź
posthog
Oficjalne

designing-email-templates

Author, save, and edit email templates in the PostHog workflows library — compose email design JSON with Liquid personalization and create and round-trip-edit templates over MCP. Use when asked to design, build, update, or fix an email template for workflows, broadcasts, or campaigns.

Instalacje / 1h
1
GitHub Stars
96
Aktualizacja
31 paź
posthog
Oficjalne

exploring-llm-evaluations

Investigate AI observability evaluations — hog (deterministic code-based), llmjudge (LLM-prompt-based), and sentiment (user-message sentiment). Find existing evaluations, inspect their configuration, run them against specific generations, query individual results, and set up scheduled reports on an evaluation. Use when the user asks to debug why an evaluation is failing, surface common failure modes, compare results across filters, dry-run a Hog evaluator, prototype a new LLM-judge prompt, inspe

Instalacje / 1h
1
GitHub Stars
96
Aktualizacja
31 paź