[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-posthog-assessing-heatmaps":3,"mdc-l4t6vy-key":41,"related-repo-posthog-assessing-heatmaps":1090,"related-org-posthog-assessing-heatmaps":1200},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":36,"sourceUrl":39,"mdContent":40},"assessing-heatmaps","analyze page heatmaps and suggest improvements","Assesses what a page's heatmap is telling you and recommends concrete changes. Pulls click \u002F rageclick \u002F scroll-depth data for a URL, names the hot elements by cross-referencing autocapture events on the same page, and can create a saved heatmap the user opens in PostHog, then summarizes the behavior and proposes improvements.\nTRIGGER when: user asks what a heatmap shows, why people aren't clicking something, where users rage-click, how far they scroll, what to change on a page based on heatmap\u002Fclick data, or to 'analyze\u002Fassess\u002Freview the heatmap' for a URL.\nDO NOT TRIGGER when: the user only wants to create a saved heatmap screenshot with no analysis (use heatmaps-saved-create directly), or is asking about session replay in general (use investigating-replay).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"posthog","PostHog","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fposthog.png",[12,14,17,20,23],{"name":9,"slug":8,"type":13},"tag",{"name":15,"slug":16,"type":13},"Product Management","product-management",{"name":18,"slug":19,"type":13},"Analytics","analytics",{"name":21,"slug":22,"type":13},"Frontend","frontend",{"name":24,"slug":25,"type":13},"UX Design","ux-design",59,"https:\u002F\u002Fgithub.com\u002FPostHog\u002Fai-plugin","2026-06-05T07:40:43.37798",null,11,[32,33,34,35],"claude-code-plugin","codex-plugin","cursor-plugin","gemini-cli-extension",{"repoUrl":27,"stars":26,"forks":30,"topics":37,"description":38},[32,33,34,35],"Official PostHog plugin for Claude Code, Cursor, Gemini, Codex and other AI coding tools","https:\u002F\u002Fgithub.com\u002FPostHog\u002Fai-plugin\u002Ftree\u002FHEAD\u002Fskills\u002Fassessing-heatmaps","---\nname: assessing-heatmaps\ndescription: \"Assesses what a page's heatmap is telling you and recommends concrete changes. Pulls click \u002F rageclick \u002F scroll-depth data for a URL, names the hot elements by cross-referencing autocapture events on the same page, and can create a saved heatmap the user opens in PostHog, then summarizes the behavior and proposes improvements.\\nTRIGGER when: user asks what a heatmap shows, why people aren't clicking something, where users rage-click, how far they scroll, what to change on a page based on heatmap\u002Fclick data, or to 'analyze\u002Fassess\u002Freview the heatmap' for a URL.\\nDO NOT TRIGGER when: the user only wants to create a saved heatmap screenshot with no analysis (use heatmaps-saved-create directly), or is asking about session replay in general (use investigating-replay).\"\n---\n\n# Assessing heatmaps\n\nA heatmap answers \"where do people interact with this page?\" — clicks, rage clicks, mouse movement, and how\nfar down they scroll. The data is pure geometry: `pointer_relative_x` (0..1 across the viewport), `pointer_y`\n(absolute pixels down the page), and a count per spot. **It does not know what was clicked.** Turning\n\"lots of clicks at (0.5, 220)\" into \"lots of clicks on the Pricing nav link\" is the whole job, and it comes\nfrom cross-referencing autocapture on the same URL.\n\n## Core principle: coordinates + meaning\n\nYou can't see the page — there is no screenshot in your context. A good assessment fuses two sources and\nleans on autocapture to supply the layout\u002Fidentity you can't see:\n\n1. **Heatmap data** — where interactions land and how far people scroll (`heatmaps-list`).\n2. **Autocapture** — what element sits under the hot spots, by element text \u002F selector on the same page. This\n   is what turns coordinates into meaning; without it you only have dots.\n\nWhen the user wants to _see_ the heatmap, create a saved heatmap (Step 4) — that renders the page with the\ndata overlaid for them to open in PostHog. You reason from the data; they look at the picture.\n\n## The flow\n\n### Step 1: Pin the page and window\n\nYou need an exact `url_exact` (one page) or a `url_pattern` (regex, to aggregate across query strings). Confirm\nthe URL with the user if ambiguous. Default to the last 7 days; widen to 30 if volume is low. Heatmap data is\nretained for 90 days.\n\n### Step 2: Pull the data\n\nCall `heatmaps-list` once per signal you care about (or query the `heatmaps` table directly via SQL — see the\nquerying-posthog-data skill, `models-heatmaps`):\n\n- `type: \"click\"` — the primary \"what draws attention\" map.\n- `type: \"rageclick\"` — repeated frustrated clicks. **The single strongest \"something is broken or\n  misleading\" signal.** Any meaningful rageclick cluster deserves a callout.\n- `type: \"scrolldepth\"` — how far people get. Use it to find the fold and spot CTAs that sit below where most\n  people ever scroll.\n\nUse `aggregation: \"unique_visitors\"` when you care about how many people (not how many clicks); `total_count`\nexaggerates a few heavy clickers.\n\nClick results come back **hottest-first** and are capped at `limit` (default 500). A busy page can have\nthousands of distinct coordinates, so the default page plus the `fold` summary is almost always enough — the\nhottest points are what analysis turns on. Don't ask for everything: raise `limit` or page with `offset` only\nwhen you specifically need more, and check `has_more` to know the list was truncated. `scrolldepth` ignores\n`limit` and always returns every bucket.\n\n### Step 2b: Above the fold — read the `fold` summary\n\nFor the click types, `heatmaps-list` returns a `fold` object alongside `results`:\n\n- `pct_below_fold` — share of non-fixed interactions that landed **below the user's initial viewport** (they\n  had to scroll to reach them). This is one of the highest-value findings: content people actively click that\n  sits below the fold is a prime candidate to move up.\n- `below_fold_count` \u002F `total_count` — the raw counts behind the percentage (fixed-position elements are\n  excluded, since they're always on screen).\n- `median_viewport_height` — the typical fold line in CSS pixels, to recommend against.\n\nReport it concretely, e.g. \"the fold is ~600px for most visitors, yet 35% of clicks land below it, so users\nscroll before interacting — that content is a candidate for the first screen.\" **Segment by device** with\n`viewport_width_min`\u002F`viewport_width_max` (desktop and mobile have very different folds) and read `fold` per\nband rather than blending them.\n\nNeed a distribution rather than a single percentage (e.g. clicks bucketed by how far below the fold)? Drop to\nSQL on the raw `heatmaps` table, which has `y` and `viewport_height` in the same scaled units — see the\nquerying-posthog-data skill, `models-heatmaps`.\n\n### Step 3: Name the hot elements (autocapture overlap)\n\nFor each notable cluster, find what's actually there. Query autocapture on the same URL — either via the\n`exploring-autocapture-events` skill or directly:\n\n```sql\nSELECT properties.$el_text AS text, count() AS clicks\nFROM events\nWHERE event = '$autocapture'\n  AND properties.$current_url = 'https:\u002F\u002Fexample.com\u002Fpricing'\n  AND timestamp >= now() - INTERVAL 7 DAY\nGROUP BY text\nORDER BY clicks DESC\nLIMIT 25\n```\n\n`elements_chain` gives the selector\u002FDOM path when you need to disambiguate two elements with the same text.\nMatch autocapture's top elements to the heatmap's hot coordinates: clicks concentrated on something that is\n**not** a link or button (plain text, an image, a disabled control) is a classic \"users expect this to be\nclickable\" finding.\n\n### Step 4: Give the user a heatmap to look at (optional)\n\nYou can't see the page, but the user can. When a visual would help them follow your findings, create a saved\nheatmap so they can open the rendered page with the data overlaid in PostHog:\n\n1. `heatmaps-saved-create` with the page `url` (type defaults to `screenshot`). This enqueues a headless\n   render — it is asynchronous. Pass `widths` matching the viewport band you analyzed in Step 2.\n2. Poll `heatmaps-saved-get` (by the returned `short_id`) until `status` is `completed`, then tell the user\n   it's ready to view in PostHog.\n\nThis is for the human's benefit — your own reasoning still comes from the Step 2 data and the Step 3\nautocapture identity, not from the picture.\n\n### Step 5: Drill into hotspots (when you need the \"why\")\n\nFor a surprising cluster, `heatmaps-events` returns the individual sessions behind specific `points`. Hand the\nsession IDs to the `investigating-replay` skill to watch what people actually did.\n\n### Step 6: Summarize and recommend\n\nProduce a short, concrete report:\n\n- **What the heatmap shows** — top engaged elements, dead zones, scroll reach, and the above\u002Fbelow-the-fold\n  click split (e.g. \"viewport is ~600px for most visitors, yet 35% of clicks land below it\").\n- **Problems**, ranked by signal strength — rage-click clusters first, then clicks on non-interactive\n  elements, then important CTAs sitting below the scroll cliff, then ignored primary actions.\n- **Recommendations** tied to evidence — move\u002Fraise a CTA above the fold, make a clicked-but-dead element a\n  real link, cut competing elements near a rage-click cluster, etc. Every recommendation should cite the\n  signal it came from.\n\n## Reading the signals\n\n| Signal                             | Likely meaning                                                                    | Typical recommendation                                         |\n| ---------------------------------- | --------------------------------------------------------------------------------- | -------------------------------------------------------------- |\n| Rage clicks on an element          | Broken, slow, or looks-clickable-but-isn't                                        | Fix the handler, add feedback, or make it actually interactive |\n| Many clicks on non-link text\u002Fimage | Users expect it to be clickable                                                   | Make it a link\u002Fbutton, or remove the affordance                |\n| Primary CTA gets few clicks        | Buried, low-contrast, or out-competed                                             | Raise it, increase contrast, reduce nearby noise               |\n| Scroll cliff before key content    | Content\u002FCTA is below where people stop                                            | Move it up or add a reason to scroll                           |\n| High % of clicks below the fold    | Engaged content sits below the initial viewport — users scroll before interacting | Move the most-clicked elements onto the first screen           |\n| Hot clicks on nav, cold body       | Page isn't delivering; people bail to nav                                         | Re-evaluate the page's core content                            |\n\n## Gotchas\n\n- **Heatmaps must be opted in** (`Team.heatmaps_opt_in`). If `heatmaps-list` returns nothing for a page that\n  clearly gets traffic, capture may be off or the URL is wrong — check both before concluding \"no\n  engagement\".\n- **Coordinates are scaled** by a factor of 16 in storage; the API already returns CSS-pixel `pointer_y` and\n  relative x, so use the API\u002Ftool values directly rather than the raw table columns.\n- **You can't see the screenshot.** The saved-heatmap render is for the user to open in PostHog; don't claim\n  to have looked at the page. Ground every layout claim in autocapture identity + coordinates, not vision.\n- **Saved-heatmap rendering is async.** After `heatmaps-saved-create`, poll `heatmaps-saved-get` until\n  `status` is `completed` before telling the user it's viewable. Only `screenshot`-type heatmaps render an\n  image; `iframe` and `recording` types do not.\n- **Mind the viewport.** A desktop click map and a mobile one are different pages' worth of behavior — filter\n  with `viewport_width_min`\u002F`viewport_width_max` rather than blending them.\n",{"data":42,"body":43},{"name":4,"description":6},{"type":44,"children":45},"root",[46,54,85,92,97,130,143,149,156,177,183,211,255,276,342,355,382,432,467,502,508,521,606,624,630,635,710,715,721,750,756,761,794,800,941,947,1084],{"type":47,"tag":48,"props":49,"children":50},"element","h1",{"id":4},[51],{"type":52,"value":53},"text","Assessing heatmaps",{"type":47,"tag":55,"props":56,"children":57},"p",{},[58,60,67,69,75,77,83],{"type":52,"value":59},"A heatmap answers \"where do people interact with this page?\" — clicks, rage clicks, mouse movement, and how\nfar down they scroll. The data is pure geometry: ",{"type":47,"tag":61,"props":62,"children":64},"code",{"className":63},[],[65],{"type":52,"value":66},"pointer_relative_x",{"type":52,"value":68}," (0..1 across the viewport), ",{"type":47,"tag":61,"props":70,"children":72},{"className":71},[],[73],{"type":52,"value":74},"pointer_y",{"type":52,"value":76},"\n(absolute pixels down the page), and a count per spot. ",{"type":47,"tag":78,"props":79,"children":80},"strong",{},[81],{"type":52,"value":82},"It does not know what was clicked.",{"type":52,"value":84}," Turning\n\"lots of clicks at (0.5, 220)\" into \"lots of clicks on the Pricing nav link\" is the whole job, and it comes\nfrom cross-referencing autocapture on the same URL.",{"type":47,"tag":86,"props":87,"children":89},"h2",{"id":88},"core-principle-coordinates-meaning",[90],{"type":52,"value":91},"Core principle: coordinates + meaning",{"type":47,"tag":55,"props":93,"children":94},{},[95],{"type":52,"value":96},"You can't see the page — there is no screenshot in your context. A good assessment fuses two sources and\nleans on autocapture to supply the layout\u002Fidentity you can't see:",{"type":47,"tag":98,"props":99,"children":100},"ol",{},[101,120],{"type":47,"tag":102,"props":103,"children":104},"li",{},[105,110,112,118],{"type":47,"tag":78,"props":106,"children":107},{},[108],{"type":52,"value":109},"Heatmap data",{"type":52,"value":111}," — where interactions land and how far people scroll (",{"type":47,"tag":61,"props":113,"children":115},{"className":114},[],[116],{"type":52,"value":117},"heatmaps-list",{"type":52,"value":119},").",{"type":47,"tag":102,"props":121,"children":122},{},[123,128],{"type":47,"tag":78,"props":124,"children":125},{},[126],{"type":52,"value":127},"Autocapture",{"type":52,"value":129}," — what element sits under the hot spots, by element text \u002F selector on the same page. This\nis what turns coordinates into meaning; without it you only have dots.",{"type":47,"tag":55,"props":131,"children":132},{},[133,135,141],{"type":52,"value":134},"When the user wants to ",{"type":47,"tag":136,"props":137,"children":138},"em",{},[139],{"type":52,"value":140},"see",{"type":52,"value":142}," the heatmap, create a saved heatmap (Step 4) — that renders the page with the\ndata overlaid for them to open in PostHog. You reason from the data; they look at the picture.",{"type":47,"tag":86,"props":144,"children":146},{"id":145},"the-flow",[147],{"type":52,"value":148},"The flow",{"type":47,"tag":150,"props":151,"children":153},"h3",{"id":152},"step-1-pin-the-page-and-window",[154],{"type":52,"value":155},"Step 1: Pin the page and window",{"type":47,"tag":55,"props":157,"children":158},{},[159,161,167,169,175],{"type":52,"value":160},"You need an exact ",{"type":47,"tag":61,"props":162,"children":164},{"className":163},[],[165],{"type":52,"value":166},"url_exact",{"type":52,"value":168}," (one page) or a ",{"type":47,"tag":61,"props":170,"children":172},{"className":171},[],[173],{"type":52,"value":174},"url_pattern",{"type":52,"value":176}," (regex, to aggregate across query strings). Confirm\nthe URL with the user if ambiguous. Default to the last 7 days; widen to 30 if volume is low. Heatmap data is\nretained for 90 days.",{"type":47,"tag":150,"props":178,"children":180},{"id":179},"step-2-pull-the-data",[181],{"type":52,"value":182},"Step 2: Pull the data",{"type":47,"tag":55,"props":184,"children":185},{},[186,188,193,195,201,203,209],{"type":52,"value":187},"Call ",{"type":47,"tag":61,"props":189,"children":191},{"className":190},[],[192],{"type":52,"value":117},{"type":52,"value":194}," once per signal you care about (or query the ",{"type":47,"tag":61,"props":196,"children":198},{"className":197},[],[199],{"type":52,"value":200},"heatmaps",{"type":52,"value":202}," table directly via SQL — see the\nquerying-posthog-data skill, ",{"type":47,"tag":61,"props":204,"children":206},{"className":205},[],[207],{"type":52,"value":208},"models-heatmaps",{"type":52,"value":210},"):",{"type":47,"tag":212,"props":213,"children":214},"ul",{},[215,226,244],{"type":47,"tag":102,"props":216,"children":217},{},[218,224],{"type":47,"tag":61,"props":219,"children":221},{"className":220},[],[222],{"type":52,"value":223},"type: \"click\"",{"type":52,"value":225}," — the primary \"what draws attention\" map.",{"type":47,"tag":102,"props":227,"children":228},{},[229,235,237,242],{"type":47,"tag":61,"props":230,"children":232},{"className":231},[],[233],{"type":52,"value":234},"type: \"rageclick\"",{"type":52,"value":236}," — repeated frustrated clicks. ",{"type":47,"tag":78,"props":238,"children":239},{},[240],{"type":52,"value":241},"The single strongest \"something is broken or\nmisleading\" signal.",{"type":52,"value":243}," Any meaningful rageclick cluster deserves a callout.",{"type":47,"tag":102,"props":245,"children":246},{},[247,253],{"type":47,"tag":61,"props":248,"children":250},{"className":249},[],[251],{"type":52,"value":252},"type: \"scrolldepth\"",{"type":52,"value":254}," — how far people get. Use it to find the fold and spot CTAs that sit below where most\npeople ever scroll.",{"type":47,"tag":55,"props":256,"children":257},{},[258,260,266,268,274],{"type":52,"value":259},"Use ",{"type":47,"tag":61,"props":261,"children":263},{"className":262},[],[264],{"type":52,"value":265},"aggregation: \"unique_visitors\"",{"type":52,"value":267}," when you care about how many people (not how many clicks); ",{"type":47,"tag":61,"props":269,"children":271},{"className":270},[],[272],{"type":52,"value":273},"total_count",{"type":52,"value":275},"\nexaggerates a few heavy clickers.",{"type":47,"tag":55,"props":277,"children":278},{},[279,281,286,288,294,296,302,304,309,311,317,319,325,327,333,335,340],{"type":52,"value":280},"Click results come back ",{"type":47,"tag":78,"props":282,"children":283},{},[284],{"type":52,"value":285},"hottest-first",{"type":52,"value":287}," and are capped at ",{"type":47,"tag":61,"props":289,"children":291},{"className":290},[],[292],{"type":52,"value":293},"limit",{"type":52,"value":295}," (default 500). A busy page can have\nthousands of distinct coordinates, so the default page plus the ",{"type":47,"tag":61,"props":297,"children":299},{"className":298},[],[300],{"type":52,"value":301},"fold",{"type":52,"value":303}," summary is almost always enough — the\nhottest points are what analysis turns on. Don't ask for everything: raise ",{"type":47,"tag":61,"props":305,"children":307},{"className":306},[],[308],{"type":52,"value":293},{"type":52,"value":310}," or page with ",{"type":47,"tag":61,"props":312,"children":314},{"className":313},[],[315],{"type":52,"value":316},"offset",{"type":52,"value":318}," only\nwhen you specifically need more, and check ",{"type":47,"tag":61,"props":320,"children":322},{"className":321},[],[323],{"type":52,"value":324},"has_more",{"type":52,"value":326}," to know the list was truncated. ",{"type":47,"tag":61,"props":328,"children":330},{"className":329},[],[331],{"type":52,"value":332},"scrolldepth",{"type":52,"value":334}," ignores\n",{"type":47,"tag":61,"props":336,"children":338},{"className":337},[],[339],{"type":52,"value":293},{"type":52,"value":341}," and always returns every bucket.",{"type":47,"tag":150,"props":343,"children":345},{"id":344},"step-2b-above-the-fold-read-the-fold-summary",[346,348,353],{"type":52,"value":347},"Step 2b: Above the fold — read the ",{"type":47,"tag":61,"props":349,"children":351},{"className":350},[],[352],{"type":52,"value":301},{"type":52,"value":354}," summary",{"type":47,"tag":55,"props":356,"children":357},{},[358,360,365,367,372,374,380],{"type":52,"value":359},"For the click types, ",{"type":47,"tag":61,"props":361,"children":363},{"className":362},[],[364],{"type":52,"value":117},{"type":52,"value":366}," returns a ",{"type":47,"tag":61,"props":368,"children":370},{"className":369},[],[371],{"type":52,"value":301},{"type":52,"value":373}," object alongside ",{"type":47,"tag":61,"props":375,"children":377},{"className":376},[],[378],{"type":52,"value":379},"results",{"type":52,"value":381},":",{"type":47,"tag":212,"props":383,"children":384},{},[385,403,421],{"type":47,"tag":102,"props":386,"children":387},{},[388,394,396,401],{"type":47,"tag":61,"props":389,"children":391},{"className":390},[],[392],{"type":52,"value":393},"pct_below_fold",{"type":52,"value":395}," — share of non-fixed interactions that landed ",{"type":47,"tag":78,"props":397,"children":398},{},[399],{"type":52,"value":400},"below the user's initial viewport",{"type":52,"value":402}," (they\nhad to scroll to reach them). This is one of the highest-value findings: content people actively click that\nsits below the fold is a prime candidate to move up.",{"type":47,"tag":102,"props":404,"children":405},{},[406,412,414,419],{"type":47,"tag":61,"props":407,"children":409},{"className":408},[],[410],{"type":52,"value":411},"below_fold_count",{"type":52,"value":413}," \u002F ",{"type":47,"tag":61,"props":415,"children":417},{"className":416},[],[418],{"type":52,"value":273},{"type":52,"value":420}," — the raw counts behind the percentage (fixed-position elements are\nexcluded, since they're always on screen).",{"type":47,"tag":102,"props":422,"children":423},{},[424,430],{"type":47,"tag":61,"props":425,"children":427},{"className":426},[],[428],{"type":52,"value":429},"median_viewport_height",{"type":52,"value":431}," — the typical fold line in CSS pixels, to recommend against.",{"type":47,"tag":55,"props":433,"children":434},{},[435,437,442,444,450,452,458,460,465],{"type":52,"value":436},"Report it concretely, e.g. \"the fold is ~600px for most visitors, yet 35% of clicks land below it, so users\nscroll before interacting — that content is a candidate for the first screen.\" ",{"type":47,"tag":78,"props":438,"children":439},{},[440],{"type":52,"value":441},"Segment by device",{"type":52,"value":443}," with\n",{"type":47,"tag":61,"props":445,"children":447},{"className":446},[],[448],{"type":52,"value":449},"viewport_width_min",{"type":52,"value":451},"\u002F",{"type":47,"tag":61,"props":453,"children":455},{"className":454},[],[456],{"type":52,"value":457},"viewport_width_max",{"type":52,"value":459}," (desktop and mobile have very different folds) and read ",{"type":47,"tag":61,"props":461,"children":463},{"className":462},[],[464],{"type":52,"value":301},{"type":52,"value":466}," per\nband rather than blending them.",{"type":47,"tag":55,"props":468,"children":469},{},[470,472,477,479,485,487,493,495,500],{"type":52,"value":471},"Need a distribution rather than a single percentage (e.g. clicks bucketed by how far below the fold)? Drop to\nSQL on the raw ",{"type":47,"tag":61,"props":473,"children":475},{"className":474},[],[476],{"type":52,"value":200},{"type":52,"value":478}," table, which has ",{"type":47,"tag":61,"props":480,"children":482},{"className":481},[],[483],{"type":52,"value":484},"y",{"type":52,"value":486}," and ",{"type":47,"tag":61,"props":488,"children":490},{"className":489},[],[491],{"type":52,"value":492},"viewport_height",{"type":52,"value":494}," in the same scaled units — see the\nquerying-posthog-data skill, ",{"type":47,"tag":61,"props":496,"children":498},{"className":497},[],[499],{"type":52,"value":208},{"type":52,"value":501},".",{"type":47,"tag":150,"props":503,"children":505},{"id":504},"step-3-name-the-hot-elements-autocapture-overlap",[506],{"type":52,"value":507},"Step 3: Name the hot elements (autocapture overlap)",{"type":47,"tag":55,"props":509,"children":510},{},[511,513,519],{"type":52,"value":512},"For each notable cluster, find what's actually there. Query autocapture on the same URL — either via the\n",{"type":47,"tag":61,"props":514,"children":516},{"className":515},[],[517],{"type":52,"value":518},"exploring-autocapture-events",{"type":52,"value":520}," skill or directly:",{"type":47,"tag":522,"props":523,"children":528},"pre",{"className":524,"code":525,"language":526,"meta":527,"style":527},"language-sql shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","SELECT properties.$el_text AS text, count() AS clicks\nFROM events\nWHERE event = '$autocapture'\n  AND properties.$current_url = 'https:\u002F\u002Fexample.com\u002Fpricing'\n  AND timestamp >= now() - INTERVAL 7 DAY\nGROUP BY text\nORDER BY clicks DESC\nLIMIT 25\n","sql","",[529],{"type":47,"tag":61,"props":530,"children":531},{"__ignoreMap":527},[532,543,552,561,570,579,588,597],{"type":47,"tag":533,"props":534,"children":537},"span",{"class":535,"line":536},"line",1,[538],{"type":47,"tag":533,"props":539,"children":540},{},[541],{"type":52,"value":542},"SELECT properties.$el_text AS text, count() AS clicks\n",{"type":47,"tag":533,"props":544,"children":546},{"class":535,"line":545},2,[547],{"type":47,"tag":533,"props":548,"children":549},{},[550],{"type":52,"value":551},"FROM events\n",{"type":47,"tag":533,"props":553,"children":555},{"class":535,"line":554},3,[556],{"type":47,"tag":533,"props":557,"children":558},{},[559],{"type":52,"value":560},"WHERE event = '$autocapture'\n",{"type":47,"tag":533,"props":562,"children":564},{"class":535,"line":563},4,[565],{"type":47,"tag":533,"props":566,"children":567},{},[568],{"type":52,"value":569},"  AND properties.$current_url = 'https:\u002F\u002Fexample.com\u002Fpricing'\n",{"type":47,"tag":533,"props":571,"children":573},{"class":535,"line":572},5,[574],{"type":47,"tag":533,"props":575,"children":576},{},[577],{"type":52,"value":578},"  AND timestamp >= now() - INTERVAL 7 DAY\n",{"type":47,"tag":533,"props":580,"children":582},{"class":535,"line":581},6,[583],{"type":47,"tag":533,"props":584,"children":585},{},[586],{"type":52,"value":587},"GROUP BY text\n",{"type":47,"tag":533,"props":589,"children":591},{"class":535,"line":590},7,[592],{"type":47,"tag":533,"props":593,"children":594},{},[595],{"type":52,"value":596},"ORDER BY clicks DESC\n",{"type":47,"tag":533,"props":598,"children":600},{"class":535,"line":599},8,[601],{"type":47,"tag":533,"props":602,"children":603},{},[604],{"type":52,"value":605},"LIMIT 25\n",{"type":47,"tag":55,"props":607,"children":608},{},[609,615,617,622],{"type":47,"tag":61,"props":610,"children":612},{"className":611},[],[613],{"type":52,"value":614},"elements_chain",{"type":52,"value":616}," gives the selector\u002FDOM path when you need to disambiguate two elements with the same text.\nMatch autocapture's top elements to the heatmap's hot coordinates: clicks concentrated on something that is\n",{"type":47,"tag":78,"props":618,"children":619},{},[620],{"type":52,"value":621},"not",{"type":52,"value":623}," a link or button (plain text, an image, a disabled control) is a classic \"users expect this to be\nclickable\" finding.",{"type":47,"tag":150,"props":625,"children":627},{"id":626},"step-4-give-the-user-a-heatmap-to-look-at-optional",[628],{"type":52,"value":629},"Step 4: Give the user a heatmap to look at (optional)",{"type":47,"tag":55,"props":631,"children":632},{},[633],{"type":52,"value":634},"You can't see the page, but the user can. When a visual would help them follow your findings, create a saved\nheatmap so they can open the rendered page with the data overlaid in PostHog:",{"type":47,"tag":98,"props":636,"children":637},{},[638,673],{"type":47,"tag":102,"props":639,"children":640},{},[641,647,649,655,657,663,665,671],{"type":47,"tag":61,"props":642,"children":644},{"className":643},[],[645],{"type":52,"value":646},"heatmaps-saved-create",{"type":52,"value":648}," with the page ",{"type":47,"tag":61,"props":650,"children":652},{"className":651},[],[653],{"type":52,"value":654},"url",{"type":52,"value":656}," (type defaults to ",{"type":47,"tag":61,"props":658,"children":660},{"className":659},[],[661],{"type":52,"value":662},"screenshot",{"type":52,"value":664},"). This enqueues a headless\nrender — it is asynchronous. Pass ",{"type":47,"tag":61,"props":666,"children":668},{"className":667},[],[669],{"type":52,"value":670},"widths",{"type":52,"value":672}," matching the viewport band you analyzed in Step 2.",{"type":47,"tag":102,"props":674,"children":675},{},[676,678,684,686,692,694,700,702,708],{"type":52,"value":677},"Poll ",{"type":47,"tag":61,"props":679,"children":681},{"className":680},[],[682],{"type":52,"value":683},"heatmaps-saved-get",{"type":52,"value":685}," (by the returned ",{"type":47,"tag":61,"props":687,"children":689},{"className":688},[],[690],{"type":52,"value":691},"short_id",{"type":52,"value":693},") until ",{"type":47,"tag":61,"props":695,"children":697},{"className":696},[],[698],{"type":52,"value":699},"status",{"type":52,"value":701}," is ",{"type":47,"tag":61,"props":703,"children":705},{"className":704},[],[706],{"type":52,"value":707},"completed",{"type":52,"value":709},", then tell the user\nit's ready to view in PostHog.",{"type":47,"tag":55,"props":711,"children":712},{},[713],{"type":52,"value":714},"This is for the human's benefit — your own reasoning still comes from the Step 2 data and the Step 3\nautocapture identity, not from the picture.",{"type":47,"tag":150,"props":716,"children":718},{"id":717},"step-5-drill-into-hotspots-when-you-need-the-why",[719],{"type":52,"value":720},"Step 5: Drill into hotspots (when you need the \"why\")",{"type":47,"tag":55,"props":722,"children":723},{},[724,726,732,734,740,742,748],{"type":52,"value":725},"For a surprising cluster, ",{"type":47,"tag":61,"props":727,"children":729},{"className":728},[],[730],{"type":52,"value":731},"heatmaps-events",{"type":52,"value":733}," returns the individual sessions behind specific ",{"type":47,"tag":61,"props":735,"children":737},{"className":736},[],[738],{"type":52,"value":739},"points",{"type":52,"value":741},". Hand the\nsession IDs to the ",{"type":47,"tag":61,"props":743,"children":745},{"className":744},[],[746],{"type":52,"value":747},"investigating-replay",{"type":52,"value":749}," skill to watch what people actually did.",{"type":47,"tag":150,"props":751,"children":753},{"id":752},"step-6-summarize-and-recommend",[754],{"type":52,"value":755},"Step 6: Summarize and recommend",{"type":47,"tag":55,"props":757,"children":758},{},[759],{"type":52,"value":760},"Produce a short, concrete report:",{"type":47,"tag":212,"props":762,"children":763},{},[764,774,784],{"type":47,"tag":102,"props":765,"children":766},{},[767,772],{"type":47,"tag":78,"props":768,"children":769},{},[770],{"type":52,"value":771},"What the heatmap shows",{"type":52,"value":773}," — top engaged elements, dead zones, scroll reach, and the above\u002Fbelow-the-fold\nclick split (e.g. \"viewport is ~600px for most visitors, yet 35% of clicks land below it\").",{"type":47,"tag":102,"props":775,"children":776},{},[777,782],{"type":47,"tag":78,"props":778,"children":779},{},[780],{"type":52,"value":781},"Problems",{"type":52,"value":783},", ranked by signal strength — rage-click clusters first, then clicks on non-interactive\nelements, then important CTAs sitting below the scroll cliff, then ignored primary actions.",{"type":47,"tag":102,"props":785,"children":786},{},[787,792],{"type":47,"tag":78,"props":788,"children":789},{},[790],{"type":52,"value":791},"Recommendations",{"type":52,"value":793}," tied to evidence — move\u002Fraise a CTA above the fold, make a clicked-but-dead element a\nreal link, cut competing elements near a rage-click cluster, etc. Every recommendation should cite the\nsignal it came from.",{"type":47,"tag":86,"props":795,"children":797},{"id":796},"reading-the-signals",[798],{"type":52,"value":799},"Reading the signals",{"type":47,"tag":801,"props":802,"children":803},"table",{},[804,828],{"type":47,"tag":805,"props":806,"children":807},"thead",{},[808],{"type":47,"tag":809,"props":810,"children":811},"tr",{},[812,818,823],{"type":47,"tag":813,"props":814,"children":815},"th",{},[816],{"type":52,"value":817},"Signal",{"type":47,"tag":813,"props":819,"children":820},{},[821],{"type":52,"value":822},"Likely meaning",{"type":47,"tag":813,"props":824,"children":825},{},[826],{"type":52,"value":827},"Typical recommendation",{"type":47,"tag":829,"props":830,"children":831},"tbody",{},[832,851,869,887,905,923],{"type":47,"tag":809,"props":833,"children":834},{},[835,841,846],{"type":47,"tag":836,"props":837,"children":838},"td",{},[839],{"type":52,"value":840},"Rage clicks on an element",{"type":47,"tag":836,"props":842,"children":843},{},[844],{"type":52,"value":845},"Broken, slow, or looks-clickable-but-isn't",{"type":47,"tag":836,"props":847,"children":848},{},[849],{"type":52,"value":850},"Fix the handler, add feedback, or make it actually interactive",{"type":47,"tag":809,"props":852,"children":853},{},[854,859,864],{"type":47,"tag":836,"props":855,"children":856},{},[857],{"type":52,"value":858},"Many clicks on non-link text\u002Fimage",{"type":47,"tag":836,"props":860,"children":861},{},[862],{"type":52,"value":863},"Users expect it to be clickable",{"type":47,"tag":836,"props":865,"children":866},{},[867],{"type":52,"value":868},"Make it a link\u002Fbutton, or remove the affordance",{"type":47,"tag":809,"props":870,"children":871},{},[872,877,882],{"type":47,"tag":836,"props":873,"children":874},{},[875],{"type":52,"value":876},"Primary CTA gets few clicks",{"type":47,"tag":836,"props":878,"children":879},{},[880],{"type":52,"value":881},"Buried, low-contrast, or out-competed",{"type":47,"tag":836,"props":883,"children":884},{},[885],{"type":52,"value":886},"Raise it, increase contrast, reduce nearby noise",{"type":47,"tag":809,"props":888,"children":889},{},[890,895,900],{"type":47,"tag":836,"props":891,"children":892},{},[893],{"type":52,"value":894},"Scroll cliff before key content",{"type":47,"tag":836,"props":896,"children":897},{},[898],{"type":52,"value":899},"Content\u002FCTA is below where people stop",{"type":47,"tag":836,"props":901,"children":902},{},[903],{"type":52,"value":904},"Move it up or add a reason to scroll",{"type":47,"tag":809,"props":906,"children":907},{},[908,913,918],{"type":47,"tag":836,"props":909,"children":910},{},[911],{"type":52,"value":912},"High % of clicks below the fold",{"type":47,"tag":836,"props":914,"children":915},{},[916],{"type":52,"value":917},"Engaged content sits below the initial viewport — users scroll before interacting",{"type":47,"tag":836,"props":919,"children":920},{},[921],{"type":52,"value":922},"Move the most-clicked elements onto the first screen",{"type":47,"tag":809,"props":924,"children":925},{},[926,931,936],{"type":47,"tag":836,"props":927,"children":928},{},[929],{"type":52,"value":930},"Hot clicks on nav, cold body",{"type":47,"tag":836,"props":932,"children":933},{},[934],{"type":52,"value":935},"Page isn't delivering; people bail to nav",{"type":47,"tag":836,"props":937,"children":938},{},[939],{"type":52,"value":940},"Re-evaluate the page's core content",{"type":47,"tag":86,"props":942,"children":944},{"id":943},"gotchas",[945],{"type":52,"value":946},"Gotchas",{"type":47,"tag":212,"props":948,"children":949},{},[950,975,992,1002,1061],{"type":47,"tag":102,"props":951,"children":952},{},[953,958,960,966,968,973],{"type":47,"tag":78,"props":954,"children":955},{},[956],{"type":52,"value":957},"Heatmaps must be opted in",{"type":52,"value":959}," (",{"type":47,"tag":61,"props":961,"children":963},{"className":962},[],[964],{"type":52,"value":965},"Team.heatmaps_opt_in",{"type":52,"value":967},"). If ",{"type":47,"tag":61,"props":969,"children":971},{"className":970},[],[972],{"type":52,"value":117},{"type":52,"value":974}," returns nothing for a page that\nclearly gets traffic, capture may be off or the URL is wrong — check both before concluding \"no\nengagement\".",{"type":47,"tag":102,"props":976,"children":977},{},[978,983,985,990],{"type":47,"tag":78,"props":979,"children":980},{},[981],{"type":52,"value":982},"Coordinates are scaled",{"type":52,"value":984}," by a factor of 16 in storage; the API already returns CSS-pixel ",{"type":47,"tag":61,"props":986,"children":988},{"className":987},[],[989],{"type":52,"value":74},{"type":52,"value":991}," and\nrelative x, so use the API\u002Ftool values directly rather than the raw table columns.",{"type":47,"tag":102,"props":993,"children":994},{},[995,1000],{"type":47,"tag":78,"props":996,"children":997},{},[998],{"type":52,"value":999},"You can't see the screenshot.",{"type":52,"value":1001}," The saved-heatmap render is for the user to open in PostHog; don't claim\nto have looked at the page. Ground every layout claim in autocapture identity + coordinates, not vision.",{"type":47,"tag":102,"props":1003,"children":1004},{},[1005,1010,1012,1017,1019,1024,1026,1031,1032,1037,1039,1044,1046,1052,1053,1059],{"type":47,"tag":78,"props":1006,"children":1007},{},[1008],{"type":52,"value":1009},"Saved-heatmap rendering is async.",{"type":52,"value":1011}," After ",{"type":47,"tag":61,"props":1013,"children":1015},{"className":1014},[],[1016],{"type":52,"value":646},{"type":52,"value":1018},", poll ",{"type":47,"tag":61,"props":1020,"children":1022},{"className":1021},[],[1023],{"type":52,"value":683},{"type":52,"value":1025}," until\n",{"type":47,"tag":61,"props":1027,"children":1029},{"className":1028},[],[1030],{"type":52,"value":699},{"type":52,"value":701},{"type":47,"tag":61,"props":1033,"children":1035},{"className":1034},[],[1036],{"type":52,"value":707},{"type":52,"value":1038}," before telling the user it's viewable. Only ",{"type":47,"tag":61,"props":1040,"children":1042},{"className":1041},[],[1043],{"type":52,"value":662},{"type":52,"value":1045},"-type heatmaps render an\nimage; ",{"type":47,"tag":61,"props":1047,"children":1049},{"className":1048},[],[1050],{"type":52,"value":1051},"iframe",{"type":52,"value":486},{"type":47,"tag":61,"props":1054,"children":1056},{"className":1055},[],[1057],{"type":52,"value":1058},"recording",{"type":52,"value":1060}," types do not.",{"type":47,"tag":102,"props":1062,"children":1063},{},[1064,1069,1071,1076,1077,1082],{"type":47,"tag":78,"props":1065,"children":1066},{},[1067],{"type":52,"value":1068},"Mind the viewport.",{"type":52,"value":1070}," A desktop click map and a mobile one are different pages' worth of behavior — filter\nwith ",{"type":47,"tag":61,"props":1072,"children":1074},{"className":1073},[],[1075],{"type":52,"value":449},{"type":52,"value":451},{"type":47,"tag":61,"props":1078,"children":1080},{"className":1079},[],[1081],{"type":52,"value":457},{"type":52,"value":1083}," rather than blending them.",{"type":47,"tag":1085,"props":1086,"children":1087},"style",{},[1088],{"type":52,"value":1089},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1091,"total":1199},[1092,1107,1115,1132,1149,1163,1181],{"slug":1093,"name":1093,"fn":1094,"description":1095,"org":1096,"tags":1097,"stars":26,"repoUrl":27,"updatedAt":1106},"analyzing-experiment-session-replays","analyze session replays for PostHog experiments","Analyze session replay patterns across experiment variants to understand user behavior differences. Use when the user wants to see how users interact with different experiment variants, identify usability issues, compare behavior patterns between control and test groups, or get qualitative insights to complement quantitative experiment results.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1098,1099,1102,1103],{"name":18,"slug":19,"type":13},{"name":1100,"slug":1101,"type":13},"Design","design",{"name":9,"slug":8,"type":13},{"name":1104,"slug":1105,"type":13},"User Research","user-research","2026-04-06T18:44:38.291781",{"slug":4,"name":4,"fn":5,"description":6,"org":1108,"tags":1109,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1110,1111,1112,1113,1114],{"name":18,"slug":19,"type":13},{"name":21,"slug":22,"type":13},{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13},{"name":24,"slug":25,"type":13},{"slug":1116,"name":1116,"fn":1117,"description":1118,"org":1119,"tags":1120,"stars":26,"repoUrl":27,"updatedAt":1131},"auditing-experiments-flags","audit PostHog experiments and feature flags","Audit PostHog experiments and feature flags for configuration issues, staleness, and best-practice violations. Read when the user asks to audit, health-check, or review experiments or feature flags, check flag hygiene, or verify experiment setup.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1121,1124,1127,1128],{"name":1122,"slug":1123,"type":13},"Audit","audit",{"name":1125,"slug":1126,"type":13},"Feature Flags","feature-flags",{"name":9,"slug":8,"type":13},{"name":1129,"slug":1130,"type":13},"QA","qa","2026-04-06T18:44:30.657553",{"slug":1133,"name":1133,"fn":1134,"description":1135,"org":1136,"tags":1137,"stars":26,"repoUrl":27,"updatedAt":1148},"authoring-scouts","author and edit PostHog Signals scouts","How to author, edit, and adapt PostHog Signals scouts — the scheduled agents that scan a project and write reports into the Signals inbox. Use when a user wants to customize a canonical scout for their own setup (narrow its scope, retune its thresholds, add disqualifiers), tweak a scout's schedule or dry-run posture, or write a brand-new scout from scratch for a specific use case (a custom event, a product surface no canonical scout covers), or steer a scout without editing it at all by leaving it a note. Covers the scout SKILL.md anatomy, the report contract, the dedupe + scratchpad-memory conventions, the scout-notes steering channel, the per-team skills-store path vs the canonical in-repo path, and the write-and-inspect test loop (with dry-run as an optional safety net). Trigger on \"write\u002Fedit\u002Fcustomize a signals scout\", \"new scout for X\", \"tune my scout schedule\", \"make a scout that watches \u003Cevent>\", \"leave a note for \u002F give feedback to a scout\", \"tell the scouts about X\".\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1138,1141,1144,1147],{"name":1139,"slug":1140,"type":13},"Agents","agents",{"name":1142,"slug":1143,"type":13},"Automation","automation",{"name":1145,"slug":1146,"type":13},"Observability","observability",{"name":9,"slug":8,"type":13},"2026-07-28T05:33:45.509154",{"slug":1150,"name":1150,"fn":1151,"description":1152,"org":1153,"tags":1154,"stars":26,"repoUrl":27,"updatedAt":1162},"building-a-dashboard","build and update PostHog dashboards","Build a new dashboard, or update an existing one, from a set of insights — the same job the in-app assistant does with its upsert-dashboard tool, but over MCP. Use when a user asks to create a dashboard, put several metrics\u002Fcharts together on one page, assemble a dashboard for a topic (product analytics, retention, revenue, activation, etc.), or add\u002Fremove\u002Freplace insights on a dashboard they already have. Covers deciding create vs update, reusing existing insights vs creating new ones, and using PostHog's vetted dashboard templates as reference for what a strong dashboard on a topic looks like.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1155,1156,1159],{"name":18,"slug":19,"type":13},{"name":1157,"slug":1158,"type":13},"Dashboards","dashboards",{"name":1160,"slug":1161,"type":13},"MCP","mcp","2026-07-21T06:07:38.060598",{"slug":1164,"name":1164,"fn":1165,"description":1166,"org":1167,"tags":1168,"stars":26,"repoUrl":27,"updatedAt":1180},"checking-deploy-timing","correlate PostHog deployments with GitHub commits","Determine when a PostHog code change reached a given environment by reading the hidden GIT deploy annotations in the project and correlating them with the merge commit on GitHub. Use when PostHog staff ask \"when was X deployed\", \"is my change live in the US\u002FEU yet\", \"has my PR shipped\", \"did the fix roll out to prod-us\", or otherwise want to know whether\u002Fwhen a commit, PR, or feature went out to a region. Do not answer deploy-timing questions from event\u002Fdata volume alone — that only shows when data changed, not when code shipped.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1169,1172,1175,1178,1179],{"name":1170,"slug":1171,"type":13},"Deployment","deployment",{"name":1173,"slug":1174,"type":13},"Git","git",{"name":1176,"slug":1177,"type":13},"GitHub","github",{"name":1145,"slug":1146,"type":13},{"name":9,"slug":8,"type":13},"2026-06-28T07:46:59.53536",{"slug":1182,"name":1182,"fn":1183,"description":1184,"org":1185,"tags":1186,"stars":26,"repoUrl":27,"updatedAt":1198},"choosing-trend-or-slope-view","visualize trends and growth over time","Clarify how to visualize change over a time range before building a trend. Use whenever the user asks how much something changed, grew, dropped, improved, or regressed between two points or periods — \"how much did X change from A to B\", \"before vs after\", \"start vs end\", \"week over week\", \"compare this month to last\", \"change over time\" — or mentions a \"slope chart\" \u002F \"slopegraph\". Two readings of \"change\" need different charts: the whole trend (a line, every interval) versus just the two endpoints (a slope, start vs end). Ask which they want, then render it. Not for choosing a saved insight ChartDisplayType in the insight editor.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1187,1188,1191,1194,1195],{"name":18,"slug":19,"type":13},{"name":1189,"slug":1190,"type":13},"Charts","charts",{"name":1192,"slug":1193,"type":13},"Data Visualization","data-visualization",{"name":9,"slug":8,"type":13},{"name":1196,"slug":1197,"type":13},"Reporting","reporting","2026-06-18T08:18:57.960157",56,{"items":1201,"total":1360},[1202,1217,1227,1240,1253,1268,1284,1297,1309,1324,1334,1350],{"slug":1203,"name":1203,"fn":1204,"description":1205,"org":1206,"tags":1207,"stars":1214,"repoUrl":1215,"updatedAt":1216},"analyzing-expensive-users","analyze expensive users in AI observability","Analyze the most expensive users in AI observability and explain why they cost so much. Use when the user asks about top spenders, expensive users, per-user LLM cost, user-level cost drivers, or patterns behind high AI observability spend.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1208,1209,1212,1213],{"name":18,"slug":19,"type":13},{"name":1210,"slug":1211,"type":13},"Cost Optimization","cost-optimization",{"name":1145,"slug":1146,"type":13},{"name":9,"slug":8,"type":13},35568,"https:\u002F\u002Fgithub.com\u002FPostHog\u002Fposthog","2026-07-28T05:34:11.117757",{"slug":1218,"name":1218,"fn":1219,"description":1220,"org":1221,"tags":1222,"stars":1214,"repoUrl":1215,"updatedAt":1226},"auditing-endpoints","audit PostHog project endpoints","Audit every endpoint in a PostHog project for staleness, failed materialisations, and unused materialised versions. Use when the user asks \"what endpoints can I clean up?\", \"are any of my endpoints broken?\", \"which materialised versions are still being called?\", or wants a one-shot cleanup pass over the Endpoints product. Produces a prioritised report grouped by issue type, with recommended actions but does not modify anything without explicit confirmation.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1223,1224,1225],{"name":18,"slug":19,"type":13},{"name":1122,"slug":1123,"type":13},{"name":9,"slug":8,"type":13},"2026-06-08T08:08:33.693989",{"slug":1228,"name":1228,"fn":1229,"description":1230,"org":1231,"tags":1232,"stars":1214,"repoUrl":1215,"updatedAt":1239},"auditing-warehouse-source-health","audit PostHog data warehouse source health","Audit the health of a PostHog project's data warehouse sources and syncs — find every broken or degraded source connection, sync schema, and webhook channel. Use when the user asks \"why are my imports failing?\", \"what's broken with my sources?\", \"why is my warehouse data stale?\", or wants a one-shot triage of source\u002Fsync health before deciding where to dig in. Produces a prioritized report grouped by severity, with recommended next steps. For materialized-view health use `auditing-warehouse-view-health`; for a single failing sync use `diagnosing-failed-warehouse-syncs`.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1233,1234,1237,1238],{"name":1122,"slug":1123,"type":13},{"name":1235,"slug":1236,"type":13},"Data Warehouse","data-warehouse",{"name":1145,"slug":1146,"type":13},{"name":9,"slug":8,"type":13},"2026-06-18T08:22:57.67984",{"slug":1241,"name":1241,"fn":1242,"description":1243,"org":1244,"tags":1245,"stars":1214,"repoUrl":1215,"updatedAt":1252},"auditing-warehouse-view-health","audit PostHog materialized view health","Audit the health of a PostHog project's materialized views (saved queries) — find every failed materialization and flag unused or stale materialized views that cost storage and compute. Use when the user asks \"which of my views are broken?\", \"why is this materialized view failing?\", \"are any of my views wasting compute?\", or wants a one-shot triage of view health. For source\u002Fsync health use `auditing-warehouse-source-health`.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1246,1247,1248,1251],{"name":1122,"slug":1123,"type":13},{"name":1235,"slug":1236,"type":13},{"name":1249,"slug":1250,"type":13},"Performance","performance",{"name":9,"slug":8,"type":13},"2026-06-18T08:25:10.936787",{"slug":1254,"name":1254,"fn":1255,"description":1256,"org":1257,"tags":1258,"stars":1214,"repoUrl":1215,"updatedAt":1267},"authoring-error-tracking-alerts","author PostHog error tracking alerts","Author error tracking alerts that fire when an issue is created, reopened, or starts spiking. Use when the user asks to set up error notifications, route exceptions to Slack\u002Fwebhook\u002FLinear, or evaluate which error events are worth alerting on. Covers trigger-event selection, integration choice, dedup against existing alerts, and shipping with the canonical message body shape.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1259,1262,1265,1266],{"name":1260,"slug":1261,"type":13},"Alerting","alerting",{"name":1263,"slug":1264,"type":13},"Debugging","debugging",{"name":1145,"slug":1146,"type":13},{"name":9,"slug":8,"type":13},"2026-06-18T08:24:40.318583",{"slug":1269,"name":1269,"fn":1270,"description":1271,"org":1272,"tags":1273,"stars":1214,"repoUrl":1215,"updatedAt":1283},"authoring-log-alerts","author log alerts in PostHog","Author useful, low-noise log alerts on services in a PostHog project. Use when the user asks to set up alerts for their logs, suggest alerts they should add, or evaluate whether a service is worth monitoring. Covers service triage, baseline characterisation, threshold drafting, back-testing via simulate, and shipping with a notification destination.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1274,1275,1278,1279,1282],{"name":18,"slug":19,"type":13},{"name":1276,"slug":1277,"type":13},"Monitoring","monitoring",{"name":1145,"slug":1146,"type":13},{"name":1280,"slug":1281,"type":13},"Operations","operations",{"name":9,"slug":8,"type":13},"2026-07-18T05:10:54.430898",{"slug":1285,"name":1285,"fn":1286,"description":1287,"org":1288,"tags":1289,"stars":1214,"repoUrl":1215,"updatedAt":1296},"building-workflows","build and edit PostHog workflows","Build, edit, test, enable, and monitor PostHog workflows over MCP. Author the action\u002Fedge graph so it runs and opens cleanly in the visual editor, then change drafts surgically with patch operations. Use when asked to build, set up, automate, change, fix, or debug a workflow, campaign, broadcast, drip sequence, or event-triggered automation in the workflows product.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1290,1291,1292,1293],{"name":1142,"slug":1143,"type":13},{"name":1160,"slug":1161,"type":13},{"name":9,"slug":8,"type":13},{"name":1294,"slug":1295,"type":13},"Workflow Automation","workflow-automation","2026-07-28T05:34:12.167015",{"slug":1298,"name":1298,"fn":1299,"description":1300,"org":1301,"tags":1302,"stars":1214,"repoUrl":1215,"updatedAt":1308},"check-posthog-loading","inspect PostHog SDK loading across URLs","Inspect how the PostHog JavaScript SDK is loaded across a list of URLs. Use to confirm consistent installation across pages, find pages missing the snippet, detect mismatched API keys or hosts between pages, and verify the load method (head snippet vs deferred vs array.js).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1303,1304,1305,1306,1307],{"name":18,"slug":19,"type":13},{"name":1263,"slug":1264,"type":13},{"name":21,"slug":22,"type":13},{"name":1145,"slug":1146,"type":13},{"name":9,"slug":8,"type":13},"2026-05-07T05:56:19.828048",{"slug":1310,"name":1310,"fn":1311,"description":1312,"org":1313,"tags":1314,"stars":1214,"repoUrl":1215,"updatedAt":1323},"consuming-endpoints-from-client-code","integrate PostHog endpoints into client applications","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\u002Fopenapi-ts, sending the right auth header, shaping the variables payload (HogQL code_name 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\".\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1315,1318,1319,1320],{"name":1316,"slug":1317,"type":13},"API Development","api-development",{"name":21,"slug":22,"type":13},{"name":9,"slug":8,"type":13},{"name":1321,"slug":1322,"type":13},"SDK","sdk","2026-06-08T08:08:34.929454",{"slug":1325,"name":1325,"fn":1326,"description":1327,"org":1328,"tags":1329,"stars":1214,"repoUrl":1215,"updatedAt":1333},"copying-endpoints-across-projects","copy PostHog endpoints across projects","Copy a PostHog endpoint (a saved HogQL\u002Finsight query exposed as an API route) to another project in the same organization, or duplicate it under a new name in the same project. Use when the user wants to duplicate an endpoint, promote an endpoint from staging to production, replicate an endpoint's query\u002Fvariables\u002Ffreshness config in another workspace, or clone an endpoint to iterate on it. Unlike feature flags and experiments, endpoints have NO native cross-project copy tool — this skill covers the read-then-recreate flow (endpoint-get then endpoint-create), the active-project switching it requires, name-collision checks, and the safe defaults (land unmaterialised in the target, verify with endpoint-run). Does not cover editing endpoint versions (see managing-endpoint-versions) or authoring a brand-new endpoint from scratch (see creating-an-endpoint).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1330,1331,1332],{"name":1316,"slug":1317,"type":13},{"name":1280,"slug":1281,"type":13},{"name":9,"slug":8,"type":13},"2026-07-15T05:29:58.442727",{"slug":1335,"name":1335,"fn":1336,"description":1337,"org":1338,"tags":1339,"stars":1214,"repoUrl":1215,"updatedAt":1349},"creating-ai-subscription","schedule recurring AI-generated PostHog reports","Create a recurring AI-generated PostHog report — schedule a free-text prompt to run on a cron, with the LLM-synthesized markdown delivered to email or Slack on each tick. Use when the user wants a recurring AI summary of X on any cadence (daily, weekly, monthly, yearly) rather than a one-off report. (To attach an AI summary to an existing insight\u002Fdashboard subscription instead of a free-text prompt, see `managing-subscriptions` and its `summary_enabled` option.)\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1340,1341,1344,1345,1346],{"name":1142,"slug":1143,"type":13},{"name":1342,"slug":1343,"type":13},"Email","email",{"name":9,"slug":8,"type":13},{"name":1196,"slug":1197,"type":13},{"name":1347,"slug":1348,"type":13},"Slack","slack","2026-06-09T07:32:27.935712",{"slug":1351,"name":1351,"fn":1352,"description":1353,"org":1354,"tags":1355,"stars":1214,"repoUrl":1215,"updatedAt":1359},"creating-an-endpoint","create PostHog API endpoints","Create a PostHog endpoint with the right shape on the first try — covers query kind choice, name conventions, what to expose as variables (HogQL code_name vs insight breakdown), data_freshness_seconds, and whether to materialise on day one. Use when the user says \"create an endpoint\", \"expose this query as an API\", \"turn this insight into an endpoint\", or asks for help structuring a new endpoint. Steers away from common mistakes: materialising a query with cohort breakdowns or compare mode, inline-only variables on a materialised endpoint, unbounded date ranges, ambiguous names.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1356,1357,1358],{"name":18,"slug":19,"type":13},{"name":1316,"slug":1317,"type":13},{"name":9,"slug":8,"type":13},"2026-06-08T08:08:29.624498",231]