[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-posthog-exploring-replay-vision-observations":3,"mdc-a12f4x-key":49,"related-repo-posthog-exploring-replay-vision-observations":1095,"related-org-posthog-exploring-replay-vision-observations":1195},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":44,"sourceUrl":47,"mdContent":48},"exploring-replay-vision-observations","explore Replay Vision scanner observations","Guides agents through pulling a Replay Vision scanner's observations, reading the findings, and acting on them — summarizing patterns across sessions, drilling into individual recordings, and turning real, corroborated issues into PostHog tasks, insights, or an investigating-replay hand-off.\nTRIGGER when: user wants to pull\u002Fread\u002Ftriage Replay Vision observations, asks \"what has my scanner found\", wants to act on or summarize scanner findings, turn observations into tasks\u002Fwork, or points at a \u002Freplay-vision\u002F\u003Cscanner-id> URL.\nDO NOT TRIGGER when: creating or sizing a scanner (use creating-replay-vision-scanners), running a one-off scan you don't then analyse, or authoring a signals scout.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"posthog","PostHog","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fposthog.png",[12,16,17,20],{"name":13,"slug":14,"type":15},"Observability","observability","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Replay Vision","replay-vision",{"name":21,"slug":22,"type":15},"Debugging","debugging",35568,"https:\u002F\u002Fgithub.com\u002FPostHog\u002Fposthog","2026-06-27T07:10:30.613408",null,2977,[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"ab-testing","ai-analytics","analytics","cdp","data-warehouse","experiments","feature-flags","javascript","product-analytics","python","react","session-replay","surveys","typescript","web-analytics",{"repoUrl":24,"stars":23,"forks":27,"topics":45,"description":46},[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"🦔 PostHog is an all-in-one developer platform for building successful products. We offer product analytics, web analytics, session replay, error tracking, feature flags, experimentation, surveys, data warehouse, a CDP, and an AI product assistant to help debug your code, ship features faster, and keep all your usage and customer data in one stack.","https:\u002F\u002Fgithub.com\u002FPostHog\u002Fposthog\u002Ftree\u002FHEAD\u002Fproducts\u002Freplay_vision\u002Fskills\u002Fexploring-replay-vision-observations","---\nname: exploring-replay-vision-observations\ndescription: \"Guides agents through pulling a Replay Vision scanner's observations, reading the findings, and acting on them — summarizing patterns across sessions, drilling into individual recordings, and turning real, corroborated issues into PostHog tasks, insights, or an investigating-replay hand-off.\\nTRIGGER when: user wants to pull\u002Fread\u002Ftriage Replay Vision observations, asks \\\"what has my scanner found\\\", wants to act on or summarize scanner findings, turn observations into tasks\u002Fwork, or points at a \u002Freplay-vision\u002F\u003Cscanner-id> URL.\\nDO NOT TRIGGER when: creating or sizing a scanner (use creating-replay-vision-scanners), running a one-off scan you don't then analyse, or authoring a signals scout.\"\n---\n\n# Exploring Replay Vision observations\n\nA scanner is a standing LLM probe over session recordings; each time it runs against a session it records\none **observation**. This skill is about the other half of the loop — reading what the scanners have found\nand doing something useful with it. For creating or sizing scanners, use [[creating-replay-vision-scanners]].\n\n## Mental model\n\n- **Scanner → observations.** One observation = one scan of one session. There is at most one observation\n  per `(scanner, session)`.\n- **The finding lives in `scanner_result`.** Its shape depends on the scanner's `scanner_type`, but it always\n  carries a `confidence`:\n  - `monitor` → a `verdict` (`yes` \u002F `no` \u002F `inconclusive`) plus an open-ended observation.\n  - `classifier` → one or more `tags` from the scanner's label set.\n  - `scorer` → a numeric score on the scanner's `scale`.\n  - `summarizer` → a free-text summary (optionally with facet embeddings).\n- **Only `succeeded` observations carry a finding.** Triage the rest by `status`\u002F`error_reason` (see below).\n- **Observations are LLM judgments, not ground truth.** One observation is one model's read of one session —\n  corroborate before you act on it.\n\nIf a scanner has `emits_signals: true`, its observations also feed the Signals pipeline and may surface as\nInbox **signal reports** (clusters of related findings). When the user's intent is \"work the reports\", that's\nthe inbox path — see _Acting on findings_ below.\n\n## Step 1 — Anchor on the scanner\n\nIf the user gave a `\u002Fproject\u002F\u003Cid>\u002Freplay-vision\u002F\u003Cscanner-id>` URL, that path segment is the scanner ID.\nOtherwise list them with `vision-scanners-list` and pick the relevant one.\n\nThen call `vision-scanners-get` to read its configuration **before** reading results — the `scanner_type` and\n`scanner_config.prompt` tell you how to interpret `scanner_result` (a `verdict` field only makes sense once you\nknow it's a monitor; a score only means something against the scorer's `scale`).\n\n## Step 2 — Pull the observations\n\nPick the axis that matches the question:\n\n- **What has this scanner found, over time?** → `vision-scanners-observations-list` (the workhorse). Filter to\n  `status=succeeded` to get only sessions with a finding, then narrow by `verdict` (monitors) or `tags`\n  (classifiers). Scorers aren't filtered by score — rank them with `order_by=-result_score` instead. Use\n  `order_by` (e.g. `-result_score`, `-completed_at`) to surface the strongest hits first.\n- **What did every scanner find about one session?** → `vision-observations-list` (the `session_id` query\n  parameter is REQUIRED). Use this while investigating a single recording.\n- **The full detail of one finding** → `vision-scanners-observations-get` or `vision-observations-retrieve` —\n  returns the frozen `scanner_snapshot` (config at run time) and the complete `scanner_result`, including any\n  event citations that link the finding back to specific events in the recording.\n\nTriage `status` so you don't mistake a non-result for \"nothing wrong\":\n\n| status                | meaning                                                       | typical `error_reason`                                                                                   |\n| --------------------- | ------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------- |\n| `succeeded`           | has a `scanner_result`                                        | —                                                                                                        |\n| `ineligible`          | session couldn't be analysed — a normal outcome, not an error | `too_short`, `no_recording`, `too_inactive`, `too_long`, `no_events`                                     |\n| `failed`              | the scan errored                                              | `provider_rejected`, `validation_failed`, `rasterization_failed`, `provider_transient`, `internal_error` |\n| `pending` \u002F `running` | still in flight                                               | —                                                                                                        |\n\nA scanner that looks like it \"found nothing\" is often producing mostly `ineligible` observations — check the\nmix before concluding.\n\n## Step 3 — Read the findings\n\n- **Monitors:** focus on `verdict: yes`; treat `inconclusive` as a weak signal. The observation text is the\n  substance.\n- **Classifiers:** group by `tags` to see the distribution of what's happening across sessions.\n- **Scorers:** look at the tails (highest\u002Flowest scores), not just the average.\n- **Summarizers:** read for recurring themes across summaries.\n\nWeight by `confidence`, and don't over-index on a single observation. To understand a specific hit, take its\n`session_id` and either cross-reference other scanners (`vision-observations-list`) or drill into the actual\nrecording with the [[investigating-replay]] skill and the session-recording MCP tools.\n\nTo test a scanner's lens against a specific session that doesn't have an observation yet, trigger one on demand\nwith `vision-scanners-scan-session` — it's async (minutes; rasterising the recording + the LLM call are slow)\nand, like all observations, runs at most once per `(scanner, session)`.\n\n## Step 4 — Act on the findings\n\nMatch the action to the user's intent, and **corroborate before you create work**:\n\n- **Summarize a pattern.** Report the finding back with the numbers and a few representative `session_id`s\n  (e.g. \"12 of 40 succeeded observations flagged checkout confusion; sessions A, B, C\"). Cite, don't assert.\n- **Make it trackable.** When a finding is corroborated across several sessions (not one low-confidence\n  hit), capture it durably with the tools that exist: create an `insight` or `notebook` to track its\n  frequency, bundle the supporting recordings into a session-recording playlist so a human can watch the\n  evidence, and add an `annotation` if it marks a regression. There is **no MCP tool to open a PostHog\n  task directly** — to route a finding into tracked work, use the Inbox path below (for signal-emitting\n  scanners) or hand the summary to a human or coding agent to act on. Group by distinct issue, not per\n  observation.\n- **Work the Inbox.** If the scanner emits signals, its findings may already be clustered into signal reports —\n  read and act on those with `inbox-reports-list` + `inbox-report-artefacts-list` (the report's work log is the\n  evidence). See the [[inbox-exploration]] skill; that path also records your work against the report.\n\nThe discipline that matters: a single observation is one model's judgment on one recording. Confirm a finding\nreproduces across observations (or against the raw recording) before turning it into a task, an alert, or a\nclaim — the same rigor the signals pipeline applies before it promotes observations to a report.\n\n## Gotchas\n\n- **Only `succeeded` observations have a `scanner_result`** — everything else is triage metadata.\n- **`ineligible` ≠ `failed`.** Ineligible is a normal terminal outcome (e.g. the recording was too short), not\n  a bug to chase.\n- **One observation per `(scanner, session)`** — re-scanning a session that already has any observation\n  (even ineligible\u002Ffailed) is a no-op.\n- **Findings are snapshotted.** Each observation keeps the `scanner_snapshot` it ran under, so older\n  observations may reflect a previous prompt\u002Fconfig (`scanner_version`).\n- **Quota is shared.** On-demand scans count against the org's monthly budget — check `vision-quota-retrieve`\n  before triggering a batch of them.\n",{"data":50,"body":51},{"name":4,"description":6},{"type":52,"children":53},"root",[54,62,76,83,277,305,311,332,388,394,399,539,551,750,762,768,833,859,878,884,896,982,987,993],{"type":55,"tag":56,"props":57,"children":58},"element","h1",{"id":4},[59],{"type":60,"value":61},"text","Exploring Replay Vision observations",{"type":55,"tag":63,"props":64,"children":65},"p",{},[66,68,74],{"type":60,"value":67},"A scanner is a standing LLM probe over session recordings; each time it runs against a session it records\none ",{"type":55,"tag":69,"props":70,"children":71},"strong",{},[72],{"type":60,"value":73},"observation",{"type":60,"value":75},". This skill is about the other half of the loop — reading what the scanners have found\nand doing something useful with it. For creating or sizing scanners, use [[creating-replay-vision-scanners]].",{"type":55,"tag":77,"props":78,"children":80},"h2",{"id":79},"mental-model",[81],{"type":60,"value":82},"Mental model",{"type":55,"tag":84,"props":85,"children":86},"ul",{},[87,107,233,267],{"type":55,"tag":88,"props":89,"children":90},"li",{},[91,96,98,105],{"type":55,"tag":69,"props":92,"children":93},{},[94],{"type":60,"value":95},"Scanner → observations.",{"type":60,"value":97}," One observation = one scan of one session. There is at most one observation\nper ",{"type":55,"tag":99,"props":100,"children":102},"code",{"className":101},[],[103],{"type":60,"value":104},"(scanner, session)",{"type":60,"value":106},".",{"type":55,"tag":88,"props":108,"children":109},{},[110,122,124,130,132,138,140],{"type":55,"tag":69,"props":111,"children":112},{},[113,115,121],{"type":60,"value":114},"The finding lives in ",{"type":55,"tag":99,"props":116,"children":118},{"className":117},[],[119],{"type":60,"value":120},"scanner_result",{"type":60,"value":106},{"type":60,"value":123}," Its shape depends on the scanner's ",{"type":55,"tag":99,"props":125,"children":127},{"className":126},[],[128],{"type":60,"value":129},"scanner_type",{"type":60,"value":131},", but it always\ncarries a ",{"type":55,"tag":99,"props":133,"children":135},{"className":134},[],[136],{"type":60,"value":137},"confidence",{"type":60,"value":139},":\n",{"type":55,"tag":84,"props":141,"children":142},{},[143,185,204,222],{"type":55,"tag":88,"props":144,"children":145},{},[146,152,154,160,162,168,170,176,177,183],{"type":55,"tag":99,"props":147,"children":149},{"className":148},[],[150],{"type":60,"value":151},"monitor",{"type":60,"value":153}," → a ",{"type":55,"tag":99,"props":155,"children":157},{"className":156},[],[158],{"type":60,"value":159},"verdict",{"type":60,"value":161}," (",{"type":55,"tag":99,"props":163,"children":165},{"className":164},[],[166],{"type":60,"value":167},"yes",{"type":60,"value":169}," \u002F ",{"type":55,"tag":99,"props":171,"children":173},{"className":172},[],[174],{"type":60,"value":175},"no",{"type":60,"value":169},{"type":55,"tag":99,"props":178,"children":180},{"className":179},[],[181],{"type":60,"value":182},"inconclusive",{"type":60,"value":184},") plus an open-ended observation.",{"type":55,"tag":88,"props":186,"children":187},{},[188,194,196,202],{"type":55,"tag":99,"props":189,"children":191},{"className":190},[],[192],{"type":60,"value":193},"classifier",{"type":60,"value":195}," → one or more ",{"type":55,"tag":99,"props":197,"children":199},{"className":198},[],[200],{"type":60,"value":201},"tags",{"type":60,"value":203}," from the scanner's label set.",{"type":55,"tag":88,"props":205,"children":206},{},[207,213,215,221],{"type":55,"tag":99,"props":208,"children":210},{"className":209},[],[211],{"type":60,"value":212},"scorer",{"type":60,"value":214}," → a numeric score on the scanner's ",{"type":55,"tag":99,"props":216,"children":218},{"className":217},[],[219],{"type":60,"value":220},"scale",{"type":60,"value":106},{"type":55,"tag":88,"props":223,"children":224},{},[225,231],{"type":55,"tag":99,"props":226,"children":228},{"className":227},[],[229],{"type":60,"value":230},"summarizer",{"type":60,"value":232}," → a free-text summary (optionally with facet embeddings).",{"type":55,"tag":88,"props":234,"children":235},{},[236,249,251,257,259,265],{"type":55,"tag":69,"props":237,"children":238},{},[239,241,247],{"type":60,"value":240},"Only ",{"type":55,"tag":99,"props":242,"children":244},{"className":243},[],[245],{"type":60,"value":246},"succeeded",{"type":60,"value":248}," observations carry a finding.",{"type":60,"value":250}," Triage the rest by ",{"type":55,"tag":99,"props":252,"children":254},{"className":253},[],[255],{"type":60,"value":256},"status",{"type":60,"value":258},"\u002F",{"type":55,"tag":99,"props":260,"children":262},{"className":261},[],[263],{"type":60,"value":264},"error_reason",{"type":60,"value":266}," (see below).",{"type":55,"tag":88,"props":268,"children":269},{},[270,275],{"type":55,"tag":69,"props":271,"children":272},{},[273],{"type":60,"value":274},"Observations are LLM judgments, not ground truth.",{"type":60,"value":276}," One observation is one model's read of one session —\ncorroborate before you act on it.",{"type":55,"tag":63,"props":278,"children":279},{},[280,282,288,290,295,297,303],{"type":60,"value":281},"If a scanner has ",{"type":55,"tag":99,"props":283,"children":285},{"className":284},[],[286],{"type":60,"value":287},"emits_signals: true",{"type":60,"value":289},", its observations also feed the Signals pipeline and may surface as\nInbox ",{"type":55,"tag":69,"props":291,"children":292},{},[293],{"type":60,"value":294},"signal reports",{"type":60,"value":296}," (clusters of related findings). When the user's intent is \"work the reports\", that's\nthe inbox path — see ",{"type":55,"tag":298,"props":299,"children":300},"em",{},[301],{"type":60,"value":302},"Acting on findings",{"type":60,"value":304}," below.",{"type":55,"tag":77,"props":306,"children":308},{"id":307},"step-1-anchor-on-the-scanner",[309],{"type":60,"value":310},"Step 1 — Anchor on the scanner",{"type":55,"tag":63,"props":312,"children":313},{},[314,316,322,324,330],{"type":60,"value":315},"If the user gave a ",{"type":55,"tag":99,"props":317,"children":319},{"className":318},[],[320],{"type":60,"value":321},"\u002Fproject\u002F\u003Cid>\u002Freplay-vision\u002F\u003Cscanner-id>",{"type":60,"value":323}," URL, that path segment is the scanner ID.\nOtherwise list them with ",{"type":55,"tag":99,"props":325,"children":327},{"className":326},[],[328],{"type":60,"value":329},"vision-scanners-list",{"type":60,"value":331}," and pick the relevant one.",{"type":55,"tag":63,"props":333,"children":334},{},[335,337,343,345,350,352,357,359,365,367,372,374,379,381,386],{"type":60,"value":336},"Then call ",{"type":55,"tag":99,"props":338,"children":340},{"className":339},[],[341],{"type":60,"value":342},"vision-scanners-get",{"type":60,"value":344}," to read its configuration ",{"type":55,"tag":69,"props":346,"children":347},{},[348],{"type":60,"value":349},"before",{"type":60,"value":351}," reading results — the ",{"type":55,"tag":99,"props":353,"children":355},{"className":354},[],[356],{"type":60,"value":129},{"type":60,"value":358}," and\n",{"type":55,"tag":99,"props":360,"children":362},{"className":361},[],[363],{"type":60,"value":364},"scanner_config.prompt",{"type":60,"value":366}," tell you how to interpret ",{"type":55,"tag":99,"props":368,"children":370},{"className":369},[],[371],{"type":60,"value":120},{"type":60,"value":373}," (a ",{"type":55,"tag":99,"props":375,"children":377},{"className":376},[],[378],{"type":60,"value":159},{"type":60,"value":380}," field only makes sense once you\nknow it's a monitor; a score only means something against the scorer's ",{"type":55,"tag":99,"props":382,"children":384},{"className":383},[],[385],{"type":60,"value":220},{"type":60,"value":387},").",{"type":55,"tag":77,"props":389,"children":391},{"id":390},"step-2-pull-the-observations",[392],{"type":60,"value":393},"Step 2 — Pull the observations",{"type":55,"tag":63,"props":395,"children":396},{},[397],{"type":60,"value":398},"Pick the axis that matches the question:",{"type":55,"tag":84,"props":400,"children":401},{},[402,474,499],{"type":55,"tag":88,"props":403,"children":404},{},[405,410,412,418,420,426,428,433,435,440,442,448,450,456,458,464,466,472],{"type":55,"tag":69,"props":406,"children":407},{},[408],{"type":60,"value":409},"What has this scanner found, over time?",{"type":60,"value":411}," → ",{"type":55,"tag":99,"props":413,"children":415},{"className":414},[],[416],{"type":60,"value":417},"vision-scanners-observations-list",{"type":60,"value":419}," (the workhorse). Filter to\n",{"type":55,"tag":99,"props":421,"children":423},{"className":422},[],[424],{"type":60,"value":425},"status=succeeded",{"type":60,"value":427}," to get only sessions with a finding, then narrow by ",{"type":55,"tag":99,"props":429,"children":431},{"className":430},[],[432],{"type":60,"value":159},{"type":60,"value":434}," (monitors) or ",{"type":55,"tag":99,"props":436,"children":438},{"className":437},[],[439],{"type":60,"value":201},{"type":60,"value":441},"\n(classifiers). Scorers aren't filtered by score — rank them with ",{"type":55,"tag":99,"props":443,"children":445},{"className":444},[],[446],{"type":60,"value":447},"order_by=-result_score",{"type":60,"value":449}," instead. Use\n",{"type":55,"tag":99,"props":451,"children":453},{"className":452},[],[454],{"type":60,"value":455},"order_by",{"type":60,"value":457}," (e.g. ",{"type":55,"tag":99,"props":459,"children":461},{"className":460},[],[462],{"type":60,"value":463},"-result_score",{"type":60,"value":465},", ",{"type":55,"tag":99,"props":467,"children":469},{"className":468},[],[470],{"type":60,"value":471},"-completed_at",{"type":60,"value":473},") to surface the strongest hits first.",{"type":55,"tag":88,"props":475,"children":476},{},[477,482,483,489,491,497],{"type":55,"tag":69,"props":478,"children":479},{},[480],{"type":60,"value":481},"What did every scanner find about one session?",{"type":60,"value":411},{"type":55,"tag":99,"props":484,"children":486},{"className":485},[],[487],{"type":60,"value":488},"vision-observations-list",{"type":60,"value":490}," (the ",{"type":55,"tag":99,"props":492,"children":494},{"className":493},[],[495],{"type":60,"value":496},"session_id",{"type":60,"value":498}," query\nparameter is REQUIRED). Use this while investigating a single recording.",{"type":55,"tag":88,"props":500,"children":501},{},[502,507,508,514,516,522,524,530,532,537],{"type":55,"tag":69,"props":503,"children":504},{},[505],{"type":60,"value":506},"The full detail of one finding",{"type":60,"value":411},{"type":55,"tag":99,"props":509,"children":511},{"className":510},[],[512],{"type":60,"value":513},"vision-scanners-observations-get",{"type":60,"value":515}," or ",{"type":55,"tag":99,"props":517,"children":519},{"className":518},[],[520],{"type":60,"value":521},"vision-observations-retrieve",{"type":60,"value":523}," —\nreturns the frozen ",{"type":55,"tag":99,"props":525,"children":527},{"className":526},[],[528],{"type":60,"value":529},"scanner_snapshot",{"type":60,"value":531}," (config at run time) and the complete ",{"type":55,"tag":99,"props":533,"children":535},{"className":534},[],[536],{"type":60,"value":120},{"type":60,"value":538},", including any\nevent citations that link the finding back to specific events in the recording.",{"type":55,"tag":63,"props":540,"children":541},{},[542,544,549],{"type":60,"value":543},"Triage ",{"type":55,"tag":99,"props":545,"children":547},{"className":546},[],[548],{"type":60,"value":256},{"type":60,"value":550}," so you don't mistake a non-result for \"nothing wrong\":",{"type":55,"tag":552,"props":553,"children":554},"table",{},[555,583],{"type":55,"tag":556,"props":557,"children":558},"thead",{},[559],{"type":55,"tag":560,"props":561,"children":562},"tr",{},[563,568,573],{"type":55,"tag":564,"props":565,"children":566},"th",{},[567],{"type":60,"value":256},{"type":55,"tag":564,"props":569,"children":570},{},[571],{"type":60,"value":572},"meaning",{"type":55,"tag":564,"props":574,"children":575},{},[576,578],{"type":60,"value":577},"typical ",{"type":55,"tag":99,"props":579,"children":581},{"className":580},[],[582],{"type":60,"value":264},{"type":55,"tag":584,"props":585,"children":586},"tbody",{},[587,614,668,722],{"type":55,"tag":560,"props":588,"children":589},{},[590,599,609],{"type":55,"tag":591,"props":592,"children":593},"td",{},[594],{"type":55,"tag":99,"props":595,"children":597},{"className":596},[],[598],{"type":60,"value":246},{"type":55,"tag":591,"props":600,"children":601},{},[602,604],{"type":60,"value":603},"has a ",{"type":55,"tag":99,"props":605,"children":607},{"className":606},[],[608],{"type":60,"value":120},{"type":55,"tag":591,"props":610,"children":611},{},[612],{"type":60,"value":613},"—",{"type":55,"tag":560,"props":615,"children":616},{},[617,626,631],{"type":55,"tag":591,"props":618,"children":619},{},[620],{"type":55,"tag":99,"props":621,"children":623},{"className":622},[],[624],{"type":60,"value":625},"ineligible",{"type":55,"tag":591,"props":627,"children":628},{},[629],{"type":60,"value":630},"session couldn't be analysed — a normal outcome, not an error",{"type":55,"tag":591,"props":632,"children":633},{},[634,640,641,647,648,654,655,661,662],{"type":55,"tag":99,"props":635,"children":637},{"className":636},[],[638],{"type":60,"value":639},"too_short",{"type":60,"value":465},{"type":55,"tag":99,"props":642,"children":644},{"className":643},[],[645],{"type":60,"value":646},"no_recording",{"type":60,"value":465},{"type":55,"tag":99,"props":649,"children":651},{"className":650},[],[652],{"type":60,"value":653},"too_inactive",{"type":60,"value":465},{"type":55,"tag":99,"props":656,"children":658},{"className":657},[],[659],{"type":60,"value":660},"too_long",{"type":60,"value":465},{"type":55,"tag":99,"props":663,"children":665},{"className":664},[],[666],{"type":60,"value":667},"no_events",{"type":55,"tag":560,"props":669,"children":670},{},[671,680,685],{"type":55,"tag":591,"props":672,"children":673},{},[674],{"type":55,"tag":99,"props":675,"children":677},{"className":676},[],[678],{"type":60,"value":679},"failed",{"type":55,"tag":591,"props":681,"children":682},{},[683],{"type":60,"value":684},"the scan errored",{"type":55,"tag":591,"props":686,"children":687},{},[688,694,695,701,702,708,709,715,716],{"type":55,"tag":99,"props":689,"children":691},{"className":690},[],[692],{"type":60,"value":693},"provider_rejected",{"type":60,"value":465},{"type":55,"tag":99,"props":696,"children":698},{"className":697},[],[699],{"type":60,"value":700},"validation_failed",{"type":60,"value":465},{"type":55,"tag":99,"props":703,"children":705},{"className":704},[],[706],{"type":60,"value":707},"rasterization_failed",{"type":60,"value":465},{"type":55,"tag":99,"props":710,"children":712},{"className":711},[],[713],{"type":60,"value":714},"provider_transient",{"type":60,"value":465},{"type":55,"tag":99,"props":717,"children":719},{"className":718},[],[720],{"type":60,"value":721},"internal_error",{"type":55,"tag":560,"props":723,"children":724},{},[725,741,746],{"type":55,"tag":591,"props":726,"children":727},{},[728,734,735],{"type":55,"tag":99,"props":729,"children":731},{"className":730},[],[732],{"type":60,"value":733},"pending",{"type":60,"value":169},{"type":55,"tag":99,"props":736,"children":738},{"className":737},[],[739],{"type":60,"value":740},"running",{"type":55,"tag":591,"props":742,"children":743},{},[744],{"type":60,"value":745},"still in flight",{"type":55,"tag":591,"props":747,"children":748},{},[749],{"type":60,"value":613},{"type":55,"tag":63,"props":751,"children":752},{},[753,755,760],{"type":60,"value":754},"A scanner that looks like it \"found nothing\" is often producing mostly ",{"type":55,"tag":99,"props":756,"children":758},{"className":757},[],[759],{"type":60,"value":625},{"type":60,"value":761}," observations — check the\nmix before concluding.",{"type":55,"tag":77,"props":763,"children":765},{"id":764},"step-3-read-the-findings",[766],{"type":60,"value":767},"Step 3 — Read the findings",{"type":55,"tag":84,"props":769,"children":770},{},[771,796,813,823],{"type":55,"tag":88,"props":772,"children":773},{},[774,779,781,787,789,794],{"type":55,"tag":69,"props":775,"children":776},{},[777],{"type":60,"value":778},"Monitors:",{"type":60,"value":780}," focus on ",{"type":55,"tag":99,"props":782,"children":784},{"className":783},[],[785],{"type":60,"value":786},"verdict: yes",{"type":60,"value":788},"; treat ",{"type":55,"tag":99,"props":790,"children":792},{"className":791},[],[793],{"type":60,"value":182},{"type":60,"value":795}," as a weak signal. The observation text is the\nsubstance.",{"type":55,"tag":88,"props":797,"children":798},{},[799,804,806,811],{"type":55,"tag":69,"props":800,"children":801},{},[802],{"type":60,"value":803},"Classifiers:",{"type":60,"value":805}," group by ",{"type":55,"tag":99,"props":807,"children":809},{"className":808},[],[810],{"type":60,"value":201},{"type":60,"value":812}," to see the distribution of what's happening across sessions.",{"type":55,"tag":88,"props":814,"children":815},{},[816,821],{"type":55,"tag":69,"props":817,"children":818},{},[819],{"type":60,"value":820},"Scorers:",{"type":60,"value":822}," look at the tails (highest\u002Flowest scores), not just the average.",{"type":55,"tag":88,"props":824,"children":825},{},[826,831],{"type":55,"tag":69,"props":827,"children":828},{},[829],{"type":60,"value":830},"Summarizers:",{"type":60,"value":832}," read for recurring themes across summaries.",{"type":55,"tag":63,"props":834,"children":835},{},[836,838,843,845,850,852,857],{"type":60,"value":837},"Weight by ",{"type":55,"tag":99,"props":839,"children":841},{"className":840},[],[842],{"type":60,"value":137},{"type":60,"value":844},", and don't over-index on a single observation. To understand a specific hit, take its\n",{"type":55,"tag":99,"props":846,"children":848},{"className":847},[],[849],{"type":60,"value":496},{"type":60,"value":851}," and either cross-reference other scanners (",{"type":55,"tag":99,"props":853,"children":855},{"className":854},[],[856],{"type":60,"value":488},{"type":60,"value":858},") or drill into the actual\nrecording with the [[investigating-replay]] skill and the session-recording MCP tools.",{"type":55,"tag":63,"props":860,"children":861},{},[862,864,870,872,877],{"type":60,"value":863},"To test a scanner's lens against a specific session that doesn't have an observation yet, trigger one on demand\nwith ",{"type":55,"tag":99,"props":865,"children":867},{"className":866},[],[868],{"type":60,"value":869},"vision-scanners-scan-session",{"type":60,"value":871}," — it's async (minutes; rasterising the recording + the LLM call are slow)\nand, like all observations, runs at most once per ",{"type":55,"tag":99,"props":873,"children":875},{"className":874},[],[876],{"type":60,"value":104},{"type":60,"value":106},{"type":55,"tag":77,"props":879,"children":881},{"id":880},"step-4-act-on-the-findings",[882],{"type":60,"value":883},"Step 4 — Act on the findings",{"type":55,"tag":63,"props":885,"children":886},{},[887,889,894],{"type":60,"value":888},"Match the action to the user's intent, and ",{"type":55,"tag":69,"props":890,"children":891},{},[892],{"type":60,"value":893},"corroborate before you create work",{"type":60,"value":895},":",{"type":55,"tag":84,"props":897,"children":898},{},[899,916,956],{"type":55,"tag":88,"props":900,"children":901},{},[902,907,909,914],{"type":55,"tag":69,"props":903,"children":904},{},[905],{"type":60,"value":906},"Summarize a pattern.",{"type":60,"value":908}," Report the finding back with the numbers and a few representative ",{"type":55,"tag":99,"props":910,"children":912},{"className":911},[],[913],{"type":60,"value":496},{"type":60,"value":915},"s\n(e.g. \"12 of 40 succeeded observations flagged checkout confusion; sessions A, B, C\"). Cite, don't assert.",{"type":55,"tag":88,"props":917,"children":918},{},[919,924,926,932,933,939,941,947,949,954],{"type":55,"tag":69,"props":920,"children":921},{},[922],{"type":60,"value":923},"Make it trackable.",{"type":60,"value":925}," When a finding is corroborated across several sessions (not one low-confidence\nhit), capture it durably with the tools that exist: create an ",{"type":55,"tag":99,"props":927,"children":929},{"className":928},[],[930],{"type":60,"value":931},"insight",{"type":60,"value":515},{"type":55,"tag":99,"props":934,"children":936},{"className":935},[],[937],{"type":60,"value":938},"notebook",{"type":60,"value":940}," to track its\nfrequency, bundle the supporting recordings into a session-recording playlist so a human can watch the\nevidence, and add an ",{"type":55,"tag":99,"props":942,"children":944},{"className":943},[],[945],{"type":60,"value":946},"annotation",{"type":60,"value":948}," if it marks a regression. There is ",{"type":55,"tag":69,"props":950,"children":951},{},[952],{"type":60,"value":953},"no MCP tool to open a PostHog\ntask directly",{"type":60,"value":955}," — to route a finding into tracked work, use the Inbox path below (for signal-emitting\nscanners) or hand the summary to a human or coding agent to act on. Group by distinct issue, not per\nobservation.",{"type":55,"tag":88,"props":957,"children":958},{},[959,964,966,972,974,980],{"type":55,"tag":69,"props":960,"children":961},{},[962],{"type":60,"value":963},"Work the Inbox.",{"type":60,"value":965}," If the scanner emits signals, its findings may already be clustered into signal reports —\nread and act on those with ",{"type":55,"tag":99,"props":967,"children":969},{"className":968},[],[970],{"type":60,"value":971},"inbox-reports-list",{"type":60,"value":973}," + ",{"type":55,"tag":99,"props":975,"children":977},{"className":976},[],[978],{"type":60,"value":979},"inbox-report-artefacts-list",{"type":60,"value":981}," (the report's work log is the\nevidence). See the [[inbox-exploration]] skill; that path also records your work against the report.",{"type":55,"tag":63,"props":983,"children":984},{},[985],{"type":60,"value":986},"The discipline that matters: a single observation is one model's judgment on one recording. Confirm a finding\nreproduces across observations (or against the raw recording) before turning it into a task, an alert, or a\nclaim — the same rigor the signals pipeline applies before it promotes observations to a report.",{"type":55,"tag":77,"props":988,"children":990},{"id":989},"gotchas",[991],{"type":60,"value":992},"Gotchas",{"type":55,"tag":84,"props":994,"children":995},{},[996,1017,1038,1053,1077],{"type":55,"tag":88,"props":997,"children":998},{},[999,1015],{"type":55,"tag":69,"props":1000,"children":1001},{},[1002,1003,1008,1010],{"type":60,"value":240},{"type":55,"tag":99,"props":1004,"children":1006},{"className":1005},[],[1007],{"type":60,"value":246},{"type":60,"value":1009}," observations have a ",{"type":55,"tag":99,"props":1011,"children":1013},{"className":1012},[],[1014],{"type":60,"value":120},{"type":60,"value":1016}," — everything else is triage metadata.",{"type":55,"tag":88,"props":1018,"children":1019},{},[1020,1036],{"type":55,"tag":69,"props":1021,"children":1022},{},[1023,1028,1030,1035],{"type":55,"tag":99,"props":1024,"children":1026},{"className":1025},[],[1027],{"type":60,"value":625},{"type":60,"value":1029}," ≠ ",{"type":55,"tag":99,"props":1031,"children":1033},{"className":1032},[],[1034],{"type":60,"value":679},{"type":60,"value":106},{"type":60,"value":1037}," Ineligible is a normal terminal outcome (e.g. the recording was too short), not\na bug to chase.",{"type":55,"tag":88,"props":1039,"children":1040},{},[1041,1051],{"type":55,"tag":69,"props":1042,"children":1043},{},[1044,1046],{"type":60,"value":1045},"One observation per ",{"type":55,"tag":99,"props":1047,"children":1049},{"className":1048},[],[1050],{"type":60,"value":104},{"type":60,"value":1052}," — re-scanning a session that already has any observation\n(even ineligible\u002Ffailed) is a no-op.",{"type":55,"tag":88,"props":1054,"children":1055},{},[1056,1061,1063,1068,1070,1076],{"type":55,"tag":69,"props":1057,"children":1058},{},[1059],{"type":60,"value":1060},"Findings are snapshotted.",{"type":60,"value":1062}," Each observation keeps the ",{"type":55,"tag":99,"props":1064,"children":1066},{"className":1065},[],[1067],{"type":60,"value":529},{"type":60,"value":1069}," it ran under, so older\nobservations may reflect a previous prompt\u002Fconfig (",{"type":55,"tag":99,"props":1071,"children":1073},{"className":1072},[],[1074],{"type":60,"value":1075},"scanner_version",{"type":60,"value":387},{"type":55,"tag":88,"props":1078,"children":1079},{},[1080,1085,1087,1093],{"type":55,"tag":69,"props":1081,"children":1082},{},[1083],{"type":60,"value":1084},"Quota is shared.",{"type":60,"value":1086}," On-demand scans count against the org's monthly budget — check ",{"type":55,"tag":99,"props":1088,"children":1090},{"className":1089},[],[1091],{"type":60,"value":1092},"vision-quota-retrieve",{"type":60,"value":1094},"\nbefore triggering a batch of them.",{"items":1096,"total":1194},[1097,1111,1123,1135,1148,1161,1177],{"slug":1098,"name":1098,"fn":1099,"description":1100,"org":1101,"tags":1102,"stars":23,"repoUrl":24,"updatedAt":1110},"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},[1103,1105,1108,1109],{"name":1104,"slug":31,"type":15},"Analytics",{"name":1106,"slug":1107,"type":15},"Cost Optimization","cost-optimization",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-28T05:34:11.117757",{"slug":1112,"name":1112,"fn":1113,"description":1114,"org":1115,"tags":1116,"stars":23,"repoUrl":24,"updatedAt":1122},"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},[1117,1118,1121],{"name":1104,"slug":31,"type":15},{"name":1119,"slug":1120,"type":15},"Audit","audit",{"name":9,"slug":8,"type":15},"2026-06-08T08:08:33.693989",{"slug":1124,"name":1124,"fn":1125,"description":1126,"org":1127,"tags":1128,"stars":23,"repoUrl":24,"updatedAt":1134},"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},[1129,1130,1132,1133],{"name":1119,"slug":1120,"type":15},{"name":1131,"slug":33,"type":15},"Data Warehouse",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-06-18T08:22:57.67984",{"slug":1136,"name":1136,"fn":1137,"description":1138,"org":1139,"tags":1140,"stars":23,"repoUrl":24,"updatedAt":1147},"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},[1141,1142,1143,1146],{"name":1119,"slug":1120,"type":15},{"name":1131,"slug":33,"type":15},{"name":1144,"slug":1145,"type":15},"Performance","performance",{"name":9,"slug":8,"type":15},"2026-06-18T08:25:10.936787",{"slug":1149,"name":1149,"fn":1150,"description":1151,"org":1152,"tags":1153,"stars":23,"repoUrl":24,"updatedAt":1160},"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},[1154,1157,1158,1159],{"name":1155,"slug":1156,"type":15},"Alerting","alerting",{"name":21,"slug":22,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-06-18T08:24:40.318583",{"slug":1162,"name":1162,"fn":1163,"description":1164,"org":1165,"tags":1166,"stars":23,"repoUrl":24,"updatedAt":1176},"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},[1167,1168,1171,1172,1175],{"name":1104,"slug":31,"type":15},{"name":1169,"slug":1170,"type":15},"Monitoring","monitoring",{"name":13,"slug":14,"type":15},{"name":1173,"slug":1174,"type":15},"Operations","operations",{"name":9,"slug":8,"type":15},"2026-07-18T05:10:54.430898",{"slug":1178,"name":1178,"fn":1179,"description":1180,"org":1181,"tags":1182,"stars":23,"repoUrl":24,"updatedAt":1193},"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},[1183,1186,1189,1190],{"name":1184,"slug":1185,"type":15},"Automation","automation",{"name":1187,"slug":1188,"type":15},"MCP","mcp",{"name":9,"slug":8,"type":15},{"name":1191,"slug":1192,"type":15},"Workflow Automation","workflow-automation","2026-07-28T05:34:12.167015",61,{"items":1196,"total":1313},[1197,1204,1210,1217,1224,1231,1239,1246,1260,1275,1285,1303],{"slug":1098,"name":1098,"fn":1099,"description":1100,"org":1198,"tags":1199,"stars":23,"repoUrl":24,"updatedAt":1110},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1200,1201,1202,1203],{"name":1104,"slug":31,"type":15},{"name":1106,"slug":1107,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"slug":1112,"name":1112,"fn":1113,"description":1114,"org":1205,"tags":1206,"stars":23,"repoUrl":24,"updatedAt":1122},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1207,1208,1209],{"name":1104,"slug":31,"type":15},{"name":1119,"slug":1120,"type":15},{"name":9,"slug":8,"type":15},{"slug":1124,"name":1124,"fn":1125,"description":1126,"org":1211,"tags":1212,"stars":23,"repoUrl":24,"updatedAt":1134},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1213,1214,1215,1216],{"name":1119,"slug":1120,"type":15},{"name":1131,"slug":33,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"slug":1136,"name":1136,"fn":1137,"description":1138,"org":1218,"tags":1219,"stars":23,"repoUrl":24,"updatedAt":1147},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1220,1221,1222,1223],{"name":1119,"slug":1120,"type":15},{"name":1131,"slug":33,"type":15},{"name":1144,"slug":1145,"type":15},{"name":9,"slug":8,"type":15},{"slug":1149,"name":1149,"fn":1150,"description":1151,"org":1225,"tags":1226,"stars":23,"repoUrl":24,"updatedAt":1160},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1227,1228,1229,1230],{"name":1155,"slug":1156,"type":15},{"name":21,"slug":22,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"slug":1162,"name":1162,"fn":1163,"description":1164,"org":1232,"tags":1233,"stars":23,"repoUrl":24,"updatedAt":1176},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1234,1235,1236,1237,1238],{"name":1104,"slug":31,"type":15},{"name":1169,"slug":1170,"type":15},{"name":13,"slug":14,"type":15},{"name":1173,"slug":1174,"type":15},{"name":9,"slug":8,"type":15},{"slug":1178,"name":1178,"fn":1179,"description":1180,"org":1240,"tags":1241,"stars":23,"repoUrl":24,"updatedAt":1193},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1242,1243,1244,1245],{"name":1184,"slug":1185,"type":15},{"name":1187,"slug":1188,"type":15},{"name":9,"slug":8,"type":15},{"name":1191,"slug":1192,"type":15},{"slug":1247,"name":1247,"fn":1248,"description":1249,"org":1250,"tags":1251,"stars":23,"repoUrl":24,"updatedAt":1259},"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},[1252,1253,1254,1257,1258],{"name":1104,"slug":31,"type":15},{"name":21,"slug":22,"type":15},{"name":1255,"slug":1256,"type":15},"Frontend","frontend",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-05-07T05:56:19.828048",{"slug":1261,"name":1261,"fn":1262,"description":1263,"org":1264,"tags":1265,"stars":23,"repoUrl":24,"updatedAt":1274},"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},[1266,1269,1270,1271],{"name":1267,"slug":1268,"type":15},"API Development","api-development",{"name":1255,"slug":1256,"type":15},{"name":9,"slug":8,"type":15},{"name":1272,"slug":1273,"type":15},"SDK","sdk","2026-06-08T08:08:34.929454",{"slug":1276,"name":1276,"fn":1277,"description":1278,"org":1279,"tags":1280,"stars":23,"repoUrl":24,"updatedAt":1284},"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},[1281,1282,1283],{"name":1267,"slug":1268,"type":15},{"name":1173,"slug":1174,"type":15},{"name":9,"slug":8,"type":15},"2026-07-15T05:29:58.442727",{"slug":1286,"name":1286,"fn":1287,"description":1288,"org":1289,"tags":1290,"stars":23,"repoUrl":24,"updatedAt":1302},"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},[1291,1292,1295,1296,1299],{"name":1184,"slug":1185,"type":15},{"name":1293,"slug":1294,"type":15},"Email","email",{"name":9,"slug":8,"type":15},{"name":1297,"slug":1298,"type":15},"Reporting","reporting",{"name":1300,"slug":1301,"type":15},"Slack","slack","2026-06-09T07:32:27.935712",{"slug":1304,"name":1304,"fn":1305,"description":1306,"org":1307,"tags":1308,"stars":23,"repoUrl":24,"updatedAt":1312},"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},[1309,1310,1311],{"name":1104,"slug":31,"type":15},{"name":1267,"slug":1268,"type":15},{"name":9,"slug":8,"type":15},"2026-06-08T08:08:29.624498",231]