[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-posthog-exploring-mcp-intent-clusters":3,"mdc--200x3c-key":48,"related-repo-posthog-exploring-mcp-intent-clusters":601,"related-org-posthog-exploring-mcp-intent-clusters":702},{"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":43,"sourceUrl":46,"mdContent":47},"exploring-mcp-intent-clusters","explore PostHog MCP intent clusters","Explore PostHog MCP intent clusters — agent goals grouped by semantic similarity, with each cluster's tool distribution and error rates. Use when the user asks \"what are agents trying to do with the MCP?\", \"group the intents\", \"which goals fail most?\", \"what does each cluster route to?\", wants to recompute the clustering, or pastes an MCP analytics intent-clustering URL.\n",{"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],{"name":9,"slug":8,"type":13},"tag",{"name":15,"slug":16,"type":13},"MCP","mcp",{"name":18,"slug":19,"type":13},"Agents","agents",{"name":21,"slug":22,"type":13},"Analytics","analytics",35568,"https:\u002F\u002Fgithub.com\u002FPostHog\u002Fposthog","2026-07-18T05:10:52.449633",null,2977,[29,30,22,31,32,33,34,35,36,37,38,39,40,41,42],"ab-testing","ai-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":44,"description":45},[29,30,22,31,32,33,34,35,36,37,38,39,40,41,42],"🦔 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\u002Fmcp_analytics\u002Fskills\u002Fexploring-mcp-intent-clusters","---\nname: exploring-mcp-intent-clusters\ndescription: >\n  Explore PostHog MCP intent clusters — agent goals grouped by semantic\n  similarity, with each cluster's tool distribution and error rates. Use when the\n  user asks \"what are agents trying to do with the MCP?\", \"group the intents\",\n  \"which goals fail most?\", \"what does each cluster route to?\", wants to\n  recompute the clustering, or pastes an MCP analytics intent-clustering URL.\n---\n\n# Exploring MCP intent clusters\n\nIntent clustering takes the free-text `$mcp_intent` values agents attach to\ntheir tool calls, embeds them, and groups semantically similar goals into\nclusters. Each cluster carries its tool distribution, call counts, and error\nrates — answering \"what are people _trying_ to do, and does it work?\" rather\nthan \"which tool was called\".\n\nUnlike tool quality and sessions (which ultimately aggregate `$mcp_tool_call`),\nclustering needs embeddings and is **not expressible in SQL**. It is served by\ntwo typed tools backed by a stored snapshot.\n\n## Tools\n\n| Tool                                              | Purpose                                           |\n| ------------------------------------------------- | ------------------------------------------------- |\n| `posthog:mcp-analytics-intent-clusters-retrieve`  | Fetch the latest cluster snapshot for the project |\n| `posthog:mcp-analytics-intent-clusters-recompute` | Trigger an async recompute of the snapshot        |\n\n## Workflow: read the current clusters\n\n```json\nposthog:mcp-analytics-intent-clusters-retrieve\n{}\n```\n\nReturns a snapshot with `status`, `last_computed_at`, `computed_with` (the\nembedding model and clustering parameters), and a `clusters` array. Each cluster\nhas a `label`, `intent_count`, `call_count`, `error_count`, `error_rate_pct`,\n`routing_entropy`, a `tool_distribution` (which tools that goal routes to, with\nper-tool error rates), and `sample_intents`.\n\nRead clusters by `call_count` for \"what are agents mostly doing\", or by\n`error_rate_pct` for \"which goals are failing\" — a high error rate on a cluster\npoints at a class of agent goals the tools serve badly.\n\n`routing_entropy` is how spread-out a cluster's tool usage is: low entropy means\none goal reliably maps to one tool; high entropy means agents are casting around\nfor the right tool for that goal (often a missing-capability signal).\n\n## Workflow: handle an empty or stale snapshot\n\n- **Empty \u002F idle with no clusters** (`status: idle`, `clusters: []`): no run has\n  happened yet. Trigger one (below) and tell the user it computes in the\n  background.\n- **Stale `last_computed_at`**: offer to recompute.\n\n## Workflow: recompute\n\n```json\nposthog:mcp-analytics-intent-clusters-recompute\n{}\n```\n\nReturns immediately with `status: computing` (HTTP 202); the work runs in the\nbackground. Poll `posthog:mcp-analytics-intent-clusters-retrieve` until `status`\nreturns to `idle` (done) or `error`. Don't block waiting — tell the user to\nre-ask in a minute.\n\n## Constructing UI links\n\n- **Intent clustering**: `https:\u002F\u002Fapp.posthog.com\u002Fproject\u002F\u003Cproject_id>\u002Fmcp-analytics\u002Fintent-clustering`\n\n## Tips\n\n- Clusters are only as good as the `$mcp_intent` coverage — if few calls carry\n  an intent, clusters will be sparse; cross-check intent coverage with a quick\n  `countIf(toString(properties.$mcp_intent) != '')` over `$mcp_tool_call`\n- A cluster with high `error_rate_pct` plus high `routing_entropy` is the\n  strongest \"the tools don't serve this goal well\" signal — worth a closer look\n  at its `sample_intents` and `tool_distribution`\n- Recompute is throttled to one run at a time per project; a 202 while already\n  computing just re-confirms the in-flight run\n\n## Related skills\n\n- [`exploring-mcp-tool-quality`](..\u002Fexploring-mcp-tool-quality\u002FSKILL.md) —\n  per-tool error rates and latency\n- [`exploring-mcp-sessions`](..\u002Fexploring-mcp-sessions\u002FSKILL.md) — the individual\n  runs behind the intents\n",{"data":49,"body":50},{"name":4,"description":6},{"type":51,"children":52},"root",[53,61,84,105,112,174,180,213,309,328,338,344,389,395,417,460,466,485,491,555,561,595],{"type":54,"tag":55,"props":56,"children":57},"element","h1",{"id":4},[58],{"type":59,"value":60},"text","Exploring MCP intent clusters",{"type":54,"tag":62,"props":63,"children":64},"p",{},[65,67,74,76,82],{"type":59,"value":66},"Intent clustering takes the free-text ",{"type":54,"tag":68,"props":69,"children":71},"code",{"className":70},[],[72],{"type":59,"value":73},"$mcp_intent",{"type":59,"value":75}," values agents attach to\ntheir tool calls, embeds them, and groups semantically similar goals into\nclusters. Each cluster carries its tool distribution, call counts, and error\nrates — answering \"what are people ",{"type":54,"tag":77,"props":78,"children":79},"em",{},[80],{"type":59,"value":81},"trying",{"type":59,"value":83}," to do, and does it work?\" rather\nthan \"which tool was called\".",{"type":54,"tag":62,"props":85,"children":86},{},[87,89,95,97,103],{"type":59,"value":88},"Unlike tool quality and sessions (which ultimately aggregate ",{"type":54,"tag":68,"props":90,"children":92},{"className":91},[],[93],{"type":59,"value":94},"$mcp_tool_call",{"type":59,"value":96},"),\nclustering needs embeddings and is ",{"type":54,"tag":98,"props":99,"children":100},"strong",{},[101],{"type":59,"value":102},"not expressible in SQL",{"type":59,"value":104},". It is served by\ntwo typed tools backed by a stored snapshot.",{"type":54,"tag":106,"props":107,"children":109},"h2",{"id":108},"tools",[110],{"type":59,"value":111},"Tools",{"type":54,"tag":113,"props":114,"children":115},"table",{},[116,135],{"type":54,"tag":117,"props":118,"children":119},"thead",{},[120],{"type":54,"tag":121,"props":122,"children":123},"tr",{},[124,130],{"type":54,"tag":125,"props":126,"children":127},"th",{},[128],{"type":59,"value":129},"Tool",{"type":54,"tag":125,"props":131,"children":132},{},[133],{"type":59,"value":134},"Purpose",{"type":54,"tag":136,"props":137,"children":138},"tbody",{},[139,157],{"type":54,"tag":121,"props":140,"children":141},{},[142,152],{"type":54,"tag":143,"props":144,"children":145},"td",{},[146],{"type":54,"tag":68,"props":147,"children":149},{"className":148},[],[150],{"type":59,"value":151},"posthog:mcp-analytics-intent-clusters-retrieve",{"type":54,"tag":143,"props":153,"children":154},{},[155],{"type":59,"value":156},"Fetch the latest cluster snapshot for the project",{"type":54,"tag":121,"props":158,"children":159},{},[160,169],{"type":54,"tag":143,"props":161,"children":162},{},[163],{"type":54,"tag":68,"props":164,"children":166},{"className":165},[],[167],{"type":59,"value":168},"posthog:mcp-analytics-intent-clusters-recompute",{"type":54,"tag":143,"props":170,"children":171},{},[172],{"type":59,"value":173},"Trigger an async recompute of the snapshot",{"type":54,"tag":106,"props":175,"children":177},{"id":176},"workflow-read-the-current-clusters",[178],{"type":59,"value":179},"Workflow: read the current clusters",{"type":54,"tag":181,"props":182,"children":187},"pre",{"className":183,"code":184,"language":185,"meta":186,"style":186},"language-json shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","posthog:mcp-analytics-intent-clusters-retrieve\n{}\n","json","",[188],{"type":54,"tag":68,"props":189,"children":190},{"__ignoreMap":186},[191,203],{"type":54,"tag":192,"props":193,"children":196},"span",{"class":194,"line":195},"line",1,[197],{"type":54,"tag":192,"props":198,"children":200},{"style":199},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[201],{"type":59,"value":202},"posthog:mcp-analytics-intent-clusters-retrieve\n",{"type":54,"tag":192,"props":204,"children":206},{"class":194,"line":205},2,[207],{"type":54,"tag":192,"props":208,"children":210},{"style":209},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[211],{"type":59,"value":212},"{}\n",{"type":54,"tag":62,"props":214,"children":215},{},[216,218,224,226,232,233,239,241,247,249,255,256,262,263,269,270,276,277,283,285,291,293,299,301,307],{"type":59,"value":217},"Returns a snapshot with ",{"type":54,"tag":68,"props":219,"children":221},{"className":220},[],[222],{"type":59,"value":223},"status",{"type":59,"value":225},", ",{"type":54,"tag":68,"props":227,"children":229},{"className":228},[],[230],{"type":59,"value":231},"last_computed_at",{"type":59,"value":225},{"type":54,"tag":68,"props":234,"children":236},{"className":235},[],[237],{"type":59,"value":238},"computed_with",{"type":59,"value":240}," (the\nembedding model and clustering parameters), and a ",{"type":54,"tag":68,"props":242,"children":244},{"className":243},[],[245],{"type":59,"value":246},"clusters",{"type":59,"value":248}," array. Each cluster\nhas a ",{"type":54,"tag":68,"props":250,"children":252},{"className":251},[],[253],{"type":59,"value":254},"label",{"type":59,"value":225},{"type":54,"tag":68,"props":257,"children":259},{"className":258},[],[260],{"type":59,"value":261},"intent_count",{"type":59,"value":225},{"type":54,"tag":68,"props":264,"children":266},{"className":265},[],[267],{"type":59,"value":268},"call_count",{"type":59,"value":225},{"type":54,"tag":68,"props":271,"children":273},{"className":272},[],[274],{"type":59,"value":275},"error_count",{"type":59,"value":225},{"type":54,"tag":68,"props":278,"children":280},{"className":279},[],[281],{"type":59,"value":282},"error_rate_pct",{"type":59,"value":284},",\n",{"type":54,"tag":68,"props":286,"children":288},{"className":287},[],[289],{"type":59,"value":290},"routing_entropy",{"type":59,"value":292},", a ",{"type":54,"tag":68,"props":294,"children":296},{"className":295},[],[297],{"type":59,"value":298},"tool_distribution",{"type":59,"value":300}," (which tools that goal routes to, with\nper-tool error rates), and ",{"type":54,"tag":68,"props":302,"children":304},{"className":303},[],[305],{"type":59,"value":306},"sample_intents",{"type":59,"value":308},".",{"type":54,"tag":62,"props":310,"children":311},{},[312,314,319,321,326],{"type":59,"value":313},"Read clusters by ",{"type":54,"tag":68,"props":315,"children":317},{"className":316},[],[318],{"type":59,"value":268},{"type":59,"value":320}," for \"what are agents mostly doing\", or by\n",{"type":54,"tag":68,"props":322,"children":324},{"className":323},[],[325],{"type":59,"value":282},{"type":59,"value":327}," for \"which goals are failing\" — a high error rate on a cluster\npoints at a class of agent goals the tools serve badly.",{"type":54,"tag":62,"props":329,"children":330},{},[331,336],{"type":54,"tag":68,"props":332,"children":334},{"className":333},[],[335],{"type":59,"value":290},{"type":59,"value":337}," is how spread-out a cluster's tool usage is: low entropy means\none goal reliably maps to one tool; high entropy means agents are casting around\nfor the right tool for that goal (often a missing-capability signal).",{"type":54,"tag":106,"props":339,"children":341},{"id":340},"workflow-handle-an-empty-or-stale-snapshot",[342],{"type":59,"value":343},"Workflow: handle an empty or stale snapshot",{"type":54,"tag":345,"props":346,"children":347},"ul",{},[348,374],{"type":54,"tag":349,"props":350,"children":351},"li",{},[352,357,359,365,366,372],{"type":54,"tag":98,"props":353,"children":354},{},[355],{"type":59,"value":356},"Empty \u002F idle with no clusters",{"type":59,"value":358}," (",{"type":54,"tag":68,"props":360,"children":362},{"className":361},[],[363],{"type":59,"value":364},"status: idle",{"type":59,"value":225},{"type":54,"tag":68,"props":367,"children":369},{"className":368},[],[370],{"type":59,"value":371},"clusters: []",{"type":59,"value":373},"): no run has\nhappened yet. Trigger one (below) and tell the user it computes in the\nbackground.",{"type":54,"tag":349,"props":375,"children":376},{},[377,387],{"type":54,"tag":98,"props":378,"children":379},{},[380,382],{"type":59,"value":381},"Stale ",{"type":54,"tag":68,"props":383,"children":385},{"className":384},[],[386],{"type":59,"value":231},{"type":59,"value":388},": offer to recompute.",{"type":54,"tag":106,"props":390,"children":392},{"id":391},"workflow-recompute",[393],{"type":59,"value":394},"Workflow: recompute",{"type":54,"tag":181,"props":396,"children":398},{"className":183,"code":397,"language":185,"meta":186,"style":186},"posthog:mcp-analytics-intent-clusters-recompute\n{}\n",[399],{"type":54,"tag":68,"props":400,"children":401},{"__ignoreMap":186},[402,410],{"type":54,"tag":192,"props":403,"children":404},{"class":194,"line":195},[405],{"type":54,"tag":192,"props":406,"children":407},{"style":199},[408],{"type":59,"value":409},"posthog:mcp-analytics-intent-clusters-recompute\n",{"type":54,"tag":192,"props":411,"children":412},{"class":194,"line":205},[413],{"type":54,"tag":192,"props":414,"children":415},{"style":209},[416],{"type":59,"value":212},{"type":54,"tag":62,"props":418,"children":419},{},[420,422,428,430,435,437,442,444,450,452,458],{"type":59,"value":421},"Returns immediately with ",{"type":54,"tag":68,"props":423,"children":425},{"className":424},[],[426],{"type":59,"value":427},"status: computing",{"type":59,"value":429}," (HTTP 202); the work runs in the\nbackground. Poll ",{"type":54,"tag":68,"props":431,"children":433},{"className":432},[],[434],{"type":59,"value":151},{"type":59,"value":436}," until ",{"type":54,"tag":68,"props":438,"children":440},{"className":439},[],[441],{"type":59,"value":223},{"type":59,"value":443},"\nreturns to ",{"type":54,"tag":68,"props":445,"children":447},{"className":446},[],[448],{"type":59,"value":449},"idle",{"type":59,"value":451}," (done) or ",{"type":54,"tag":68,"props":453,"children":455},{"className":454},[],[456],{"type":59,"value":457},"error",{"type":59,"value":459},". Don't block waiting — tell the user to\nre-ask in a minute.",{"type":54,"tag":106,"props":461,"children":463},{"id":462},"constructing-ui-links",[464],{"type":59,"value":465},"Constructing UI links",{"type":54,"tag":345,"props":467,"children":468},{},[469],{"type":54,"tag":349,"props":470,"children":471},{},[472,477,479],{"type":54,"tag":98,"props":473,"children":474},{},[475],{"type":59,"value":476},"Intent clustering",{"type":59,"value":478},": ",{"type":54,"tag":68,"props":480,"children":482},{"className":481},[],[483],{"type":59,"value":484},"https:\u002F\u002Fapp.posthog.com\u002Fproject\u002F\u003Cproject_id>\u002Fmcp-analytics\u002Fintent-clustering",{"type":54,"tag":106,"props":486,"children":488},{"id":487},"tips",[489],{"type":59,"value":490},"Tips",{"type":54,"tag":345,"props":492,"children":493},{},[494,519,550],{"type":54,"tag":349,"props":495,"children":496},{},[497,499,504,506,512,514],{"type":59,"value":498},"Clusters are only as good as the ",{"type":54,"tag":68,"props":500,"children":502},{"className":501},[],[503],{"type":59,"value":73},{"type":59,"value":505}," coverage — if few calls carry\nan intent, clusters will be sparse; cross-check intent coverage with a quick\n",{"type":54,"tag":68,"props":507,"children":509},{"className":508},[],[510],{"type":59,"value":511},"countIf(toString(properties.$mcp_intent) != '')",{"type":59,"value":513}," over ",{"type":54,"tag":68,"props":515,"children":517},{"className":516},[],[518],{"type":59,"value":94},{"type":54,"tag":349,"props":520,"children":521},{},[522,524,529,531,536,538,543,545],{"type":59,"value":523},"A cluster with high ",{"type":54,"tag":68,"props":525,"children":527},{"className":526},[],[528],{"type":59,"value":282},{"type":59,"value":530}," plus high ",{"type":54,"tag":68,"props":532,"children":534},{"className":533},[],[535],{"type":59,"value":290},{"type":59,"value":537}," is the\nstrongest \"the tools don't serve this goal well\" signal — worth a closer look\nat its ",{"type":54,"tag":68,"props":539,"children":541},{"className":540},[],[542],{"type":59,"value":306},{"type":59,"value":544}," and ",{"type":54,"tag":68,"props":546,"children":548},{"className":547},[],[549],{"type":59,"value":298},{"type":54,"tag":349,"props":551,"children":552},{},[553],{"type":59,"value":554},"Recompute is throttled to one run at a time per project; a 202 while already\ncomputing just re-confirms the in-flight run",{"type":54,"tag":106,"props":556,"children":558},{"id":557},"related-skills",[559],{"type":59,"value":560},"Related skills",{"type":54,"tag":345,"props":562,"children":563},{},[564,580],{"type":54,"tag":349,"props":565,"children":566},{},[567,578],{"type":54,"tag":568,"props":569,"children":571},"a",{"href":570},"..\u002Fexploring-mcp-tool-quality\u002FSKILL.md",[572],{"type":54,"tag":68,"props":573,"children":575},{"className":574},[],[576],{"type":59,"value":577},"exploring-mcp-tool-quality",{"type":59,"value":579}," —\nper-tool error rates and latency",{"type":54,"tag":349,"props":581,"children":582},{},[583,593],{"type":54,"tag":568,"props":584,"children":586},{"href":585},"..\u002Fexploring-mcp-sessions\u002FSKILL.md",[587],{"type":54,"tag":68,"props":588,"children":590},{"className":589},[],[591],{"type":59,"value":592},"exploring-mcp-sessions",{"type":59,"value":594}," — the individual\nruns behind the intents",{"type":54,"tag":596,"props":597,"children":598},"style",{},[599],{"type":59,"value":600},"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":602,"total":701},[603,618,630,642,655,670,686],{"slug":604,"name":604,"fn":605,"description":606,"org":607,"tags":608,"stars":23,"repoUrl":24,"updatedAt":617},"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},[609,610,613,616],{"name":21,"slug":22,"type":13},{"name":611,"slug":612,"type":13},"Cost Optimization","cost-optimization",{"name":614,"slug":615,"type":13},"Observability","observability",{"name":9,"slug":8,"type":13},"2026-07-28T05:34:11.117757",{"slug":619,"name":619,"fn":620,"description":621,"org":622,"tags":623,"stars":23,"repoUrl":24,"updatedAt":629},"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},[624,625,628],{"name":21,"slug":22,"type":13},{"name":626,"slug":627,"type":13},"Audit","audit",{"name":9,"slug":8,"type":13},"2026-06-08T08:08:33.693989",{"slug":631,"name":631,"fn":632,"description":633,"org":634,"tags":635,"stars":23,"repoUrl":24,"updatedAt":641},"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},[636,637,639,640],{"name":626,"slug":627,"type":13},{"name":638,"slug":32,"type":13},"Data Warehouse",{"name":614,"slug":615,"type":13},{"name":9,"slug":8,"type":13},"2026-06-18T08:22:57.67984",{"slug":643,"name":643,"fn":644,"description":645,"org":646,"tags":647,"stars":23,"repoUrl":24,"updatedAt":654},"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},[648,649,650,653],{"name":626,"slug":627,"type":13},{"name":638,"slug":32,"type":13},{"name":651,"slug":652,"type":13},"Performance","performance",{"name":9,"slug":8,"type":13},"2026-06-18T08:25:10.936787",{"slug":656,"name":656,"fn":657,"description":658,"org":659,"tags":660,"stars":23,"repoUrl":24,"updatedAt":669},"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},[661,664,667,668],{"name":662,"slug":663,"type":13},"Alerting","alerting",{"name":665,"slug":666,"type":13},"Debugging","debugging",{"name":614,"slug":615,"type":13},{"name":9,"slug":8,"type":13},"2026-06-18T08:24:40.318583",{"slug":671,"name":671,"fn":672,"description":673,"org":674,"tags":675,"stars":23,"repoUrl":24,"updatedAt":685},"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},[676,677,680,681,684],{"name":21,"slug":22,"type":13},{"name":678,"slug":679,"type":13},"Monitoring","monitoring",{"name":614,"slug":615,"type":13},{"name":682,"slug":683,"type":13},"Operations","operations",{"name":9,"slug":8,"type":13},"2026-07-18T05:10:54.430898",{"slug":687,"name":687,"fn":688,"description":689,"org":690,"tags":691,"stars":23,"repoUrl":24,"updatedAt":700},"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},[692,695,696,697],{"name":693,"slug":694,"type":13},"Automation","automation",{"name":15,"slug":16,"type":13},{"name":9,"slug":8,"type":13},{"name":698,"slug":699,"type":13},"Workflow Automation","workflow-automation","2026-07-28T05:34:12.167015",61,{"items":703,"total":820},[704,711,717,724,731,738,746,753,767,782,792,810],{"slug":604,"name":604,"fn":605,"description":606,"org":705,"tags":706,"stars":23,"repoUrl":24,"updatedAt":617},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[707,708,709,710],{"name":21,"slug":22,"type":13},{"name":611,"slug":612,"type":13},{"name":614,"slug":615,"type":13},{"name":9,"slug":8,"type":13},{"slug":619,"name":619,"fn":620,"description":621,"org":712,"tags":713,"stars":23,"repoUrl":24,"updatedAt":629},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[714,715,716],{"name":21,"slug":22,"type":13},{"name":626,"slug":627,"type":13},{"name":9,"slug":8,"type":13},{"slug":631,"name":631,"fn":632,"description":633,"org":718,"tags":719,"stars":23,"repoUrl":24,"updatedAt":641},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[720,721,722,723],{"name":626,"slug":627,"type":13},{"name":638,"slug":32,"type":13},{"name":614,"slug":615,"type":13},{"name":9,"slug":8,"type":13},{"slug":643,"name":643,"fn":644,"description":645,"org":725,"tags":726,"stars":23,"repoUrl":24,"updatedAt":654},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[727,728,729,730],{"name":626,"slug":627,"type":13},{"name":638,"slug":32,"type":13},{"name":651,"slug":652,"type":13},{"name":9,"slug":8,"type":13},{"slug":656,"name":656,"fn":657,"description":658,"org":732,"tags":733,"stars":23,"repoUrl":24,"updatedAt":669},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[734,735,736,737],{"name":662,"slug":663,"type":13},{"name":665,"slug":666,"type":13},{"name":614,"slug":615,"type":13},{"name":9,"slug":8,"type":13},{"slug":671,"name":671,"fn":672,"description":673,"org":739,"tags":740,"stars":23,"repoUrl":24,"updatedAt":685},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[741,742,743,744,745],{"name":21,"slug":22,"type":13},{"name":678,"slug":679,"type":13},{"name":614,"slug":615,"type":13},{"name":682,"slug":683,"type":13},{"name":9,"slug":8,"type":13},{"slug":687,"name":687,"fn":688,"description":689,"org":747,"tags":748,"stars":23,"repoUrl":24,"updatedAt":700},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[749,750,751,752],{"name":693,"slug":694,"type":13},{"name":15,"slug":16,"type":13},{"name":9,"slug":8,"type":13},{"name":698,"slug":699,"type":13},{"slug":754,"name":754,"fn":755,"description":756,"org":757,"tags":758,"stars":23,"repoUrl":24,"updatedAt":766},"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},[759,760,761,764,765],{"name":21,"slug":22,"type":13},{"name":665,"slug":666,"type":13},{"name":762,"slug":763,"type":13},"Frontend","frontend",{"name":614,"slug":615,"type":13},{"name":9,"slug":8,"type":13},"2026-05-07T05:56:19.828048",{"slug":768,"name":768,"fn":769,"description":770,"org":771,"tags":772,"stars":23,"repoUrl":24,"updatedAt":781},"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},[773,776,777,778],{"name":774,"slug":775,"type":13},"API Development","api-development",{"name":762,"slug":763,"type":13},{"name":9,"slug":8,"type":13},{"name":779,"slug":780,"type":13},"SDK","sdk","2026-06-08T08:08:34.929454",{"slug":783,"name":783,"fn":784,"description":785,"org":786,"tags":787,"stars":23,"repoUrl":24,"updatedAt":791},"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},[788,789,790],{"name":774,"slug":775,"type":13},{"name":682,"slug":683,"type":13},{"name":9,"slug":8,"type":13},"2026-07-15T05:29:58.442727",{"slug":793,"name":793,"fn":794,"description":795,"org":796,"tags":797,"stars":23,"repoUrl":24,"updatedAt":809},"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},[798,799,802,803,806],{"name":693,"slug":694,"type":13},{"name":800,"slug":801,"type":13},"Email","email",{"name":9,"slug":8,"type":13},{"name":804,"slug":805,"type":13},"Reporting","reporting",{"name":807,"slug":808,"type":13},"Slack","slack","2026-06-09T07:32:27.935712",{"slug":811,"name":811,"fn":812,"description":813,"org":814,"tags":815,"stars":23,"repoUrl":24,"updatedAt":819},"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},[816,817,818],{"name":21,"slug":22,"type":13},{"name":774,"slug":775,"type":13},{"name":9,"slug":8,"type":13},"2026-06-08T08:08:29.624498",231]