[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-arize-phoenix-graphql":3,"mdc--ugq39d-key":50,"related-org-arize-phoenix-graphql":1432,"related-repo-arize-phoenix-graphql":1592},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":45,"sourceUrl":48,"mdContent":49},"phoenix-graphql","query Phoenix API with GraphQL","Write efficient GraphQL queries against the Phoenix API. Load this skill in two cases: (1) before composing any non-trivial GraphQL query yourself for data analysis (via the `phoenix-gql` bash command) — it contains schema entrypoints and patterns that eliminate the need for introspection; (2) when the user asks for help writing GraphQL queries for their own scripts, tools, or integrations against Phoenix — it covers the endpoint, authentication, and client examples.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"arize","Arize AI","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Farize.jpg","Arize-ai",[13,17,20],{"name":14,"slug":15,"type":16},"GraphQL","graphql","tag",{"name":18,"slug":19,"type":16},"Data Analysis","data-analysis",{"name":21,"slug":22,"type":16},"Analytics","analytics",10513,"https:\u002F\u002Fgithub.com\u002FArize-ai\u002Fphoenix","2026-07-12T08:08:17.163493",null,977,[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"agents","ai-monitoring","ai-observability","aiengineering","anthropic","datasets","evals","langchain","llamaindex","llm-eval","llm-evaluation","llmops","llms","openai","prompt-engineering","smolagents",{"repoUrl":24,"stars":23,"forks":27,"topics":46,"description":47},[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"AI Observability & Evaluation","https:\u002F\u002Fgithub.com\u002FArize-ai\u002Fphoenix\u002Ftree\u002FHEAD\u002Fsrc\u002Fphoenix\u002Fserver\u002Fagents\u002Fprompts\u002Fskills\u002Fphoenix-graphql","---\nname: phoenix-graphql\ndescription: >\n  Write efficient GraphQL queries against the Phoenix API. Load this skill in two cases:\n  (1) before composing any non-trivial GraphQL query yourself for data analysis (via\n  the `phoenix-gql` bash command) — it contains schema entrypoints and patterns\n  that eliminate the need for introspection; (2) when the user asks for help writing\n  GraphQL queries for their own scripts, tools, or integrations against Phoenix —\n  it covers the endpoint, authentication, and client examples.\nsummary: Answer data questions with efficient GraphQL queries, or get working GraphQL for your own scripts and integrations against the Phoenix API.\n---\n\n### Two modes\n\n- **Internal data analysis** — you are querying Phoenix yourself to answer a question. Apply the schema facts, efficiency rules, and patterns below directly.\n- **Helping the user integrate** — the user wants GraphQL queries for their own code or tools. Use the same schema facts and patterns, plus the \"External API usage\" section for endpoint, auth, and client examples. Queries you hand to the user should use variables and include pagination handling.\n\n### Entrypoints\n\nTop-level `Query` entrypoints get you to a starting entity; per-entity schema details live in the resources listed under \"Schema map\" below.\n\n- `node(id: ID!)` — global lookup for **any** entity by its Relay global id; resolve with an inline fragment, e.g. `node(id: $id) { ... on Dataset { name } }`. This is the primary way to fetch datasets, prompts, experiments, sessions, and annotations, which have **no** by-name\u002Fby-id helpers.\n- `projects(...)`, `datasets(...)`, `prompts(...)`, `evaluators(...)` → Relay connections, each with `filter`\u002F`sort` inputs to find an entity when you only have a name.\n- By-X helpers (the only ones that exist): `getProjectByName(name: String!)`, `getProjectSessionById(sessionId: String!)`, `getDatasetExampleByExternalId(datasetId: GlobalID!, externalId: String!)`, `getSpanByOtelId(spanId: String!)`, `getTraceByOtelId(traceId: String!)`. There is **no** `getDatasetByName`, `getPromptByName`, or `getExperimentById` — use `node(id:)` or a connection `filter` instead.\n- `viewer` → the authenticated `User`; `projectCount`, `datasetCount`, `promptCount` — cheap counts.\n- `compareExperiments(baseExperimentId: GlobalID!, compareExperimentIds: [GlobalID!]!, first, after, filterCondition)` → experiment comparison.\n\n### Schema map\n\nPer-entity field references and examples are split into resources. Read **only** the one(s) you need with `read_skill_resource`, after loading this skill:\n\n- `project-spans-traces` — Project aggregates and `spans`; Span and Trace fields. The starting point for most trace analysis.\n- `sessions` — ProjectSession: multi-turn session metrics, token\u002Fcost, session traces.\n- `datasets` — Dataset and DatasetExample: examples, versions, splits, labels.\n- `experiments` — Experiment and ExperimentRun: runs, aggregate metrics, comparison.\n- `prompts` — Prompt and PromptVersion: versions, templates, tags.\n- `annotations` — Span\u002FTrace\u002FExperimentRun annotation fields and how to read them.\n\n### Conventions\n\nThese apply to every entity:\n\n- **Pagination** is Relay-style: `first`\u002F`after` args; responses have `edges { node { ... } }` and `pageInfo { hasNextPage endCursor }`. Cursors are opaque strings. Some connections (e.g. `Project.spans`, `Experiment.runs`, `ProjectSession.traces`) are forward-only.\n- **IDs**: the `id` field on any node is a Relay global ID (base64 of `TypeName:rowId`) — use it with `node(id:)`. OpenTelemetry hex IDs come from `Span.spanId` and `Trace.traceId` — use those for OTel lookups and `\u002Fredirects\u002Fspans\u002F\u003CspanId>` \u002F `\u002Fredirects\u002Ftraces\u002F\u003CtraceId>` links. Note a `Span` has **no** `traceId` field; read it via the nested `trace { traceId }`. Never mix global IDs with OTel IDs.\n- **`TimeRange`** input: `{ start: DateTime, end: DateTime }` — ISO 8601 strings; `end` is exclusive; both optional.\n- **`SpanSort`** input: `{ col: SpanColumn, dir: SortDir }`, e.g. `{ col: startTime, dir: desc }`. Useful `SpanColumn` values: `startTime`, `latencyMs`, `tokenCountTotal`, `cumulativeTokenCountTotal`, `tokenCostTotal`.\n- **`filterCondition`** is a Python-like DSL string over span fields, e.g. `span_kind == 'LLM'`, `status_code == 'ERROR'`, `latency_ms > 1000`, `'timeout' in output.value`, `evals['Hallucination'].label == 'hallucinated'`, `annotations['note'].score \u003C 0.5`. Combine with `and`\u002F`or`.\n\n### Efficiency rules\n\n- **Do not run full schema introspection.** Read the relevant `Schema map` resource instead; it covers the fields and arguments for that entity. Only when a resource does not cover a field you need, introspect a single type: `{ __type(name: \"Project\") { fields { name args { name type { name kind } } } } }`.\n- **Batch independent lookups with aliases** in one query instead of multiple round trips, e.g. `p50: latencyMsQuantile(probability: 0.5) p99: latencyMsQuantile(probability: 0.99)`.\n- Select only the fields you need; keep page sizes small (10–50) and paginate only when necessary.\n- Pass values via query variables, never string interpolation.\n- Span `input`\u002F`output` payloads can be huge — request `input { truncatedValue }` (first 100 chars) when surveying; fetch `input { value }` (full payload) only for spans you intend to read closely.\n\n### Patterns\n\nTwo canonical shapes to orient you; entity-specific examples live in each resource.\n\nReach an entity and read fields via `node(id:)` + an inline fragment:\n\n```graphql\nquery GetEntity($id: ID!) {\n  node(id: $id) {\n    ... on Dataset { name exampleCount }\n  }\n}\n```\n\nBatch independent project aggregates with aliases in one round trip:\n\n```graphql\nquery Overview($name: String!, $timeRange: TimeRange) {\n  getProjectByName(name: $name) {\n    traceCount(timeRange: $timeRange)\n    p50: latencyMsQuantile(probability: 0.5, timeRange: $timeRange)\n    p99: latencyMsQuantile(probability: 0.99, timeRange: $timeRange)\n    errorCount: recordCount(timeRange: $timeRange, filterCondition: \"status_code == 'ERROR'\")\n  }\n}\n```\n\n### Execution surfaces (internal mode)\n\n- `phoenix-gql` (bash): run `phoenix-gql --help` for flags and current permissions. Use `--data-only` when piping to `jq`, `--output \u003Cfile>` for large results, `--vars '\u003Cjson>'` for variables. Mutations are allowed only when runtime permissions say so; the tool reports its permissions on every invocation.\n\n### External API usage (user-facing mode)\n\nFacts users need to call the API themselves:\n\n- **Endpoint**: `POST \u003Cphoenix-host>\u002Fgraphql` with a JSON body `{ \"query\": \"...\", \"variables\": { ... } }`. A GraphiQL IDE is served on GET at the same path.\n- **Auth**: send a Phoenix API key as a bearer token: `Authorization: Bearer \u003CAPI_KEY>`. API keys are created in Phoenix settings.\n- The GraphQL schema is primarily designed for the Phoenix UI and may change between versions; for stable programmatic access, recommend the REST API (`\u002Fv1\u002F...`) and the `arize-phoenix-client` Python \u002F `@arizeai\u002Fphoenix-client` TypeScript packages where they cover the need, and GraphQL for everything else.\n\ncurl:\n\n```bash\ncurl -s \"$PHOENIX_HOST\u002Fgraphql\" \\\n  -H \"Authorization: Bearer $PHOENIX_API_KEY\" \\\n  -H \"Content-Type: application\u002Fjson\" \\\n  -d '{\"query\": \"query($n: String!) { getProjectByName(name: $n) { traceCount } }\", \"variables\": {\"n\": \"default\"}}'\n```\n\nPython:\n\n```python\nimport httpx\n\nresp = httpx.post(\n    f\"{host}\u002Fgraphql\",\n    headers={\"Authorization\": f\"Bearer {api_key}\"},\n    json={\"query\": query, \"variables\": variables},\n)\nresp.raise_for_status()\ndata = resp.json()[\"data\"]\n```\n\nWhen handing users a query, include: the full operation with variable definitions, an example variables payload, and a note on paginating via `pageInfo { hasNextPage endCursor }` → pass `endCursor` as `after`.\n",{"data":51,"body":53},{"name":4,"description":6,"summary":52},"Answer data questions with efficient GraphQL queries, or get working GraphQL for your own scripts and integrations against the Phoenix API.",{"type":54,"children":55},"root",[56,65,91,97,112,334,340,360,436,442,447,780,786,876,882,887,899,956,961,1033,1039,1092,1098,1103,1179,1184,1312,1317,1400,1426],{"type":57,"tag":58,"props":59,"children":61},"element","h3",{"id":60},"two-modes",[62],{"type":63,"value":64},"text","Two modes",{"type":57,"tag":66,"props":67,"children":68},"ul",{},[69,81],{"type":57,"tag":70,"props":71,"children":72},"li",{},[73,79],{"type":57,"tag":74,"props":75,"children":76},"strong",{},[77],{"type":63,"value":78},"Internal data analysis",{"type":63,"value":80}," — you are querying Phoenix yourself to answer a question. Apply the schema facts, efficiency rules, and patterns below directly.",{"type":57,"tag":70,"props":82,"children":83},{},[84,89],{"type":57,"tag":74,"props":85,"children":86},{},[87],{"type":63,"value":88},"Helping the user integrate",{"type":63,"value":90}," — the user wants GraphQL queries for their own code or tools. Use the same schema facts and patterns, plus the \"External API usage\" section for endpoint, auth, and client examples. Queries you hand to the user should use variables and include pagination handling.",{"type":57,"tag":58,"props":92,"children":94},{"id":93},"entrypoints",[95],{"type":63,"value":96},"Entrypoints",{"type":57,"tag":98,"props":99,"children":100},"p",{},[101,103,110],{"type":63,"value":102},"Top-level ",{"type":57,"tag":104,"props":105,"children":107},"code",{"className":106},[],[108],{"type":63,"value":109},"Query",{"type":63,"value":111}," entrypoints get you to a starting entity; per-entity schema details live in the resources listed under \"Schema map\" below.",{"type":57,"tag":66,"props":113,"children":114},{},[115,148,197,282,323],{"type":57,"tag":70,"props":116,"children":117},{},[118,124,126,131,133,139,141,146],{"type":57,"tag":104,"props":119,"children":121},{"className":120},[],[122],{"type":63,"value":123},"node(id: ID!)",{"type":63,"value":125}," — global lookup for ",{"type":57,"tag":74,"props":127,"children":128},{},[129],{"type":63,"value":130},"any",{"type":63,"value":132}," entity by its Relay global id; resolve with an inline fragment, e.g. ",{"type":57,"tag":104,"props":134,"children":136},{"className":135},[],[137],{"type":63,"value":138},"node(id: $id) { ... on Dataset { name } }",{"type":63,"value":140},". This is the primary way to fetch datasets, prompts, experiments, sessions, and annotations, which have ",{"type":57,"tag":74,"props":142,"children":143},{},[144],{"type":63,"value":145},"no",{"type":63,"value":147}," by-name\u002Fby-id helpers.",{"type":57,"tag":70,"props":149,"children":150},{},[151,157,159,165,166,172,173,179,181,187,189,195],{"type":57,"tag":104,"props":152,"children":154},{"className":153},[],[155],{"type":63,"value":156},"projects(...)",{"type":63,"value":158},", ",{"type":57,"tag":104,"props":160,"children":162},{"className":161},[],[163],{"type":63,"value":164},"datasets(...)",{"type":63,"value":158},{"type":57,"tag":104,"props":167,"children":169},{"className":168},[],[170],{"type":63,"value":171},"prompts(...)",{"type":63,"value":158},{"type":57,"tag":104,"props":174,"children":176},{"className":175},[],[177],{"type":63,"value":178},"evaluators(...)",{"type":63,"value":180}," → Relay connections, each with ",{"type":57,"tag":104,"props":182,"children":184},{"className":183},[],[185],{"type":63,"value":186},"filter",{"type":63,"value":188},"\u002F",{"type":57,"tag":104,"props":190,"children":192},{"className":191},[],[193],{"type":63,"value":194},"sort",{"type":63,"value":196}," inputs to find an entity when you only have a name.",{"type":57,"tag":70,"props":198,"children":199},{},[200,202,208,209,215,216,222,223,229,230,236,238,242,244,250,251,257,259,265,267,273,275,280],{"type":63,"value":201},"By-X helpers (the only ones that exist): ",{"type":57,"tag":104,"props":203,"children":205},{"className":204},[],[206],{"type":63,"value":207},"getProjectByName(name: String!)",{"type":63,"value":158},{"type":57,"tag":104,"props":210,"children":212},{"className":211},[],[213],{"type":63,"value":214},"getProjectSessionById(sessionId: String!)",{"type":63,"value":158},{"type":57,"tag":104,"props":217,"children":219},{"className":218},[],[220],{"type":63,"value":221},"getDatasetExampleByExternalId(datasetId: GlobalID!, externalId: String!)",{"type":63,"value":158},{"type":57,"tag":104,"props":224,"children":226},{"className":225},[],[227],{"type":63,"value":228},"getSpanByOtelId(spanId: String!)",{"type":63,"value":158},{"type":57,"tag":104,"props":231,"children":233},{"className":232},[],[234],{"type":63,"value":235},"getTraceByOtelId(traceId: String!)",{"type":63,"value":237},". There is ",{"type":57,"tag":74,"props":239,"children":240},{},[241],{"type":63,"value":145},{"type":63,"value":243}," ",{"type":57,"tag":104,"props":245,"children":247},{"className":246},[],[248],{"type":63,"value":249},"getDatasetByName",{"type":63,"value":158},{"type":57,"tag":104,"props":252,"children":254},{"className":253},[],[255],{"type":63,"value":256},"getPromptByName",{"type":63,"value":258},", or ",{"type":57,"tag":104,"props":260,"children":262},{"className":261},[],[263],{"type":63,"value":264},"getExperimentById",{"type":63,"value":266}," — use ",{"type":57,"tag":104,"props":268,"children":270},{"className":269},[],[271],{"type":63,"value":272},"node(id:)",{"type":63,"value":274}," or a connection ",{"type":57,"tag":104,"props":276,"children":278},{"className":277},[],[279],{"type":63,"value":186},{"type":63,"value":281}," instead.",{"type":57,"tag":70,"props":283,"children":284},{},[285,291,293,299,301,307,308,314,315,321],{"type":57,"tag":104,"props":286,"children":288},{"className":287},[],[289],{"type":63,"value":290},"viewer",{"type":63,"value":292}," → the authenticated ",{"type":57,"tag":104,"props":294,"children":296},{"className":295},[],[297],{"type":63,"value":298},"User",{"type":63,"value":300},"; ",{"type":57,"tag":104,"props":302,"children":304},{"className":303},[],[305],{"type":63,"value":306},"projectCount",{"type":63,"value":158},{"type":57,"tag":104,"props":309,"children":311},{"className":310},[],[312],{"type":63,"value":313},"datasetCount",{"type":63,"value":158},{"type":57,"tag":104,"props":316,"children":318},{"className":317},[],[319],{"type":63,"value":320},"promptCount",{"type":63,"value":322}," — cheap counts.",{"type":57,"tag":70,"props":324,"children":325},{},[326,332],{"type":57,"tag":104,"props":327,"children":329},{"className":328},[],[330],{"type":63,"value":331},"compareExperiments(baseExperimentId: GlobalID!, compareExperimentIds: [GlobalID!]!, first, after, filterCondition)",{"type":63,"value":333}," → experiment comparison.",{"type":57,"tag":58,"props":335,"children":337},{"id":336},"schema-map",[338],{"type":63,"value":339},"Schema map",{"type":57,"tag":98,"props":341,"children":342},{},[343,345,350,352,358],{"type":63,"value":344},"Per-entity field references and examples are split into resources. Read ",{"type":57,"tag":74,"props":346,"children":347},{},[348],{"type":63,"value":349},"only",{"type":63,"value":351}," the one(s) you need with ",{"type":57,"tag":104,"props":353,"children":355},{"className":354},[],[356],{"type":63,"value":357},"read_skill_resource",{"type":63,"value":359},", after loading this skill:",{"type":57,"tag":66,"props":361,"children":362},{},[363,382,393,403,414,425],{"type":57,"tag":70,"props":364,"children":365},{},[366,372,374,380],{"type":57,"tag":104,"props":367,"children":369},{"className":368},[],[370],{"type":63,"value":371},"project-spans-traces",{"type":63,"value":373}," — Project aggregates and ",{"type":57,"tag":104,"props":375,"children":377},{"className":376},[],[378],{"type":63,"value":379},"spans",{"type":63,"value":381},"; Span and Trace fields. The starting point for most trace analysis.",{"type":57,"tag":70,"props":383,"children":384},{},[385,391],{"type":57,"tag":104,"props":386,"children":388},{"className":387},[],[389],{"type":63,"value":390},"sessions",{"type":63,"value":392}," — ProjectSession: multi-turn session metrics, token\u002Fcost, session traces.",{"type":57,"tag":70,"props":394,"children":395},{},[396,401],{"type":57,"tag":104,"props":397,"children":399},{"className":398},[],[400],{"type":63,"value":34},{"type":63,"value":402}," — Dataset and DatasetExample: examples, versions, splits, labels.",{"type":57,"tag":70,"props":404,"children":405},{},[406,412],{"type":57,"tag":104,"props":407,"children":409},{"className":408},[],[410],{"type":63,"value":411},"experiments",{"type":63,"value":413}," — Experiment and ExperimentRun: runs, aggregate metrics, comparison.",{"type":57,"tag":70,"props":415,"children":416},{},[417,423],{"type":57,"tag":104,"props":418,"children":420},{"className":419},[],[421],{"type":63,"value":422},"prompts",{"type":63,"value":424}," — Prompt and PromptVersion: versions, templates, tags.",{"type":57,"tag":70,"props":426,"children":427},{},[428,434],{"type":57,"tag":104,"props":429,"children":431},{"className":430},[],[432],{"type":63,"value":433},"annotations",{"type":63,"value":435}," — Span\u002FTrace\u002FExperimentRun annotation fields and how to read them.",{"type":57,"tag":58,"props":437,"children":439},{"id":438},"conventions",[440],{"type":63,"value":441},"Conventions",{"type":57,"tag":98,"props":443,"children":444},{},[445],{"type":63,"value":446},"These apply to every entity:",{"type":57,"tag":66,"props":448,"children":449},{},[450,513,606,636,709],{"type":57,"tag":70,"props":451,"children":452},{},[453,458,460,466,467,473,475,481,483,489,491,497,498,504,505,511],{"type":57,"tag":74,"props":454,"children":455},{},[456],{"type":63,"value":457},"Pagination",{"type":63,"value":459}," is Relay-style: ",{"type":57,"tag":104,"props":461,"children":463},{"className":462},[],[464],{"type":63,"value":465},"first",{"type":63,"value":188},{"type":57,"tag":104,"props":468,"children":470},{"className":469},[],[471],{"type":63,"value":472},"after",{"type":63,"value":474}," args; responses have ",{"type":57,"tag":104,"props":476,"children":478},{"className":477},[],[479],{"type":63,"value":480},"edges { node { ... } }",{"type":63,"value":482}," and ",{"type":57,"tag":104,"props":484,"children":486},{"className":485},[],[487],{"type":63,"value":488},"pageInfo { hasNextPage endCursor }",{"type":63,"value":490},". Cursors are opaque strings. Some connections (e.g. ",{"type":57,"tag":104,"props":492,"children":494},{"className":493},[],[495],{"type":63,"value":496},"Project.spans",{"type":63,"value":158},{"type":57,"tag":104,"props":499,"children":501},{"className":500},[],[502],{"type":63,"value":503},"Experiment.runs",{"type":63,"value":158},{"type":57,"tag":104,"props":506,"children":508},{"className":507},[],[509],{"type":63,"value":510},"ProjectSession.traces",{"type":63,"value":512},") are forward-only.",{"type":57,"tag":70,"props":514,"children":515},{},[516,521,523,529,531,537,539,544,546,552,553,559,561,567,569,575,577,583,585,589,590,596,598,604],{"type":57,"tag":74,"props":517,"children":518},{},[519],{"type":63,"value":520},"IDs",{"type":63,"value":522},": the ",{"type":57,"tag":104,"props":524,"children":526},{"className":525},[],[527],{"type":63,"value":528},"id",{"type":63,"value":530}," field on any node is a Relay global ID (base64 of ",{"type":57,"tag":104,"props":532,"children":534},{"className":533},[],[535],{"type":63,"value":536},"TypeName:rowId",{"type":63,"value":538},") — use it with ",{"type":57,"tag":104,"props":540,"children":542},{"className":541},[],[543],{"type":63,"value":272},{"type":63,"value":545},". OpenTelemetry hex IDs come from ",{"type":57,"tag":104,"props":547,"children":549},{"className":548},[],[550],{"type":63,"value":551},"Span.spanId",{"type":63,"value":482},{"type":57,"tag":104,"props":554,"children":556},{"className":555},[],[557],{"type":63,"value":558},"Trace.traceId",{"type":63,"value":560}," — use those for OTel lookups and ",{"type":57,"tag":104,"props":562,"children":564},{"className":563},[],[565],{"type":63,"value":566},"\u002Fredirects\u002Fspans\u002F\u003CspanId>",{"type":63,"value":568}," \u002F ",{"type":57,"tag":104,"props":570,"children":572},{"className":571},[],[573],{"type":63,"value":574},"\u002Fredirects\u002Ftraces\u002F\u003CtraceId>",{"type":63,"value":576}," links. Note a ",{"type":57,"tag":104,"props":578,"children":580},{"className":579},[],[581],{"type":63,"value":582},"Span",{"type":63,"value":584}," has ",{"type":57,"tag":74,"props":586,"children":587},{},[588],{"type":63,"value":145},{"type":63,"value":243},{"type":57,"tag":104,"props":591,"children":593},{"className":592},[],[594],{"type":63,"value":595},"traceId",{"type":63,"value":597}," field; read it via the nested ",{"type":57,"tag":104,"props":599,"children":601},{"className":600},[],[602],{"type":63,"value":603},"trace { traceId }",{"type":63,"value":605},". Never mix global IDs with OTel IDs.",{"type":57,"tag":70,"props":607,"children":608},{},[609,618,620,626,628,634],{"type":57,"tag":74,"props":610,"children":611},{},[612],{"type":57,"tag":104,"props":613,"children":615},{"className":614},[],[616],{"type":63,"value":617},"TimeRange",{"type":63,"value":619}," input: ",{"type":57,"tag":104,"props":621,"children":623},{"className":622},[],[624],{"type":63,"value":625},"{ start: DateTime, end: DateTime }",{"type":63,"value":627}," — ISO 8601 strings; ",{"type":57,"tag":104,"props":629,"children":631},{"className":630},[],[632],{"type":63,"value":633},"end",{"type":63,"value":635}," is exclusive; both optional.",{"type":57,"tag":70,"props":637,"children":638},{},[639,648,649,655,657,663,665,671,673,679,680,686,687,693,694,700,701,707],{"type":57,"tag":74,"props":640,"children":641},{},[642],{"type":57,"tag":104,"props":643,"children":645},{"className":644},[],[646],{"type":63,"value":647},"SpanSort",{"type":63,"value":619},{"type":57,"tag":104,"props":650,"children":652},{"className":651},[],[653],{"type":63,"value":654},"{ col: SpanColumn, dir: SortDir }",{"type":63,"value":656},", e.g. ",{"type":57,"tag":104,"props":658,"children":660},{"className":659},[],[661],{"type":63,"value":662},"{ col: startTime, dir: desc }",{"type":63,"value":664},". Useful ",{"type":57,"tag":104,"props":666,"children":668},{"className":667},[],[669],{"type":63,"value":670},"SpanColumn",{"type":63,"value":672}," values: ",{"type":57,"tag":104,"props":674,"children":676},{"className":675},[],[677],{"type":63,"value":678},"startTime",{"type":63,"value":158},{"type":57,"tag":104,"props":681,"children":683},{"className":682},[],[684],{"type":63,"value":685},"latencyMs",{"type":63,"value":158},{"type":57,"tag":104,"props":688,"children":690},{"className":689},[],[691],{"type":63,"value":692},"tokenCountTotal",{"type":63,"value":158},{"type":57,"tag":104,"props":695,"children":697},{"className":696},[],[698],{"type":63,"value":699},"cumulativeTokenCountTotal",{"type":63,"value":158},{"type":57,"tag":104,"props":702,"children":704},{"className":703},[],[705],{"type":63,"value":706},"tokenCostTotal",{"type":63,"value":708},".",{"type":57,"tag":70,"props":710,"children":711},{},[712,721,723,729,730,736,737,743,744,750,751,757,758,764,766,772,773,779],{"type":57,"tag":74,"props":713,"children":714},{},[715],{"type":57,"tag":104,"props":716,"children":718},{"className":717},[],[719],{"type":63,"value":720},"filterCondition",{"type":63,"value":722}," is a Python-like DSL string over span fields, e.g. ",{"type":57,"tag":104,"props":724,"children":726},{"className":725},[],[727],{"type":63,"value":728},"span_kind == 'LLM'",{"type":63,"value":158},{"type":57,"tag":104,"props":731,"children":733},{"className":732},[],[734],{"type":63,"value":735},"status_code == 'ERROR'",{"type":63,"value":158},{"type":57,"tag":104,"props":738,"children":740},{"className":739},[],[741],{"type":63,"value":742},"latency_ms > 1000",{"type":63,"value":158},{"type":57,"tag":104,"props":745,"children":747},{"className":746},[],[748],{"type":63,"value":749},"'timeout' in output.value",{"type":63,"value":158},{"type":57,"tag":104,"props":752,"children":754},{"className":753},[],[755],{"type":63,"value":756},"evals['Hallucination'].label == 'hallucinated'",{"type":63,"value":158},{"type":57,"tag":104,"props":759,"children":761},{"className":760},[],[762],{"type":63,"value":763},"annotations['note'].score \u003C 0.5",{"type":63,"value":765},". Combine with ",{"type":57,"tag":104,"props":767,"children":769},{"className":768},[],[770],{"type":63,"value":771},"and",{"type":63,"value":188},{"type":57,"tag":104,"props":774,"children":776},{"className":775},[],[777],{"type":63,"value":778},"or",{"type":63,"value":708},{"type":57,"tag":58,"props":781,"children":783},{"id":782},"efficiency-rules",[784],{"type":63,"value":785},"Efficiency rules",{"type":57,"tag":66,"props":787,"children":788},{},[789,813,830,835,840],{"type":57,"tag":70,"props":790,"children":791},{},[792,797,799,804,806,812],{"type":57,"tag":74,"props":793,"children":794},{},[795],{"type":63,"value":796},"Do not run full schema introspection.",{"type":63,"value":798}," Read the relevant ",{"type":57,"tag":104,"props":800,"children":802},{"className":801},[],[803],{"type":63,"value":339},{"type":63,"value":805}," resource instead; it covers the fields and arguments for that entity. Only when a resource does not cover a field you need, introspect a single type: ",{"type":57,"tag":104,"props":807,"children":809},{"className":808},[],[810],{"type":63,"value":811},"{ __type(name: \"Project\") { fields { name args { name type { name kind } } } } }",{"type":63,"value":708},{"type":57,"tag":70,"props":814,"children":815},{},[816,821,823,829],{"type":57,"tag":74,"props":817,"children":818},{},[819],{"type":63,"value":820},"Batch independent lookups with aliases",{"type":63,"value":822}," in one query instead of multiple round trips, e.g. ",{"type":57,"tag":104,"props":824,"children":826},{"className":825},[],[827],{"type":63,"value":828},"p50: latencyMsQuantile(probability: 0.5) p99: latencyMsQuantile(probability: 0.99)",{"type":63,"value":708},{"type":57,"tag":70,"props":831,"children":832},{},[833],{"type":63,"value":834},"Select only the fields you need; keep page sizes small (10–50) and paginate only when necessary.",{"type":57,"tag":70,"props":836,"children":837},{},[838],{"type":63,"value":839},"Pass values via query variables, never string interpolation.",{"type":57,"tag":70,"props":841,"children":842},{},[843,845,851,852,858,860,866,868,874],{"type":63,"value":844},"Span ",{"type":57,"tag":104,"props":846,"children":848},{"className":847},[],[849],{"type":63,"value":850},"input",{"type":63,"value":188},{"type":57,"tag":104,"props":853,"children":855},{"className":854},[],[856],{"type":63,"value":857},"output",{"type":63,"value":859}," payloads can be huge — request ",{"type":57,"tag":104,"props":861,"children":863},{"className":862},[],[864],{"type":63,"value":865},"input { truncatedValue }",{"type":63,"value":867}," (first 100 chars) when surveying; fetch ",{"type":57,"tag":104,"props":869,"children":871},{"className":870},[],[872],{"type":63,"value":873},"input { value }",{"type":63,"value":875}," (full payload) only for spans you intend to read closely.",{"type":57,"tag":58,"props":877,"children":879},{"id":878},"patterns",[880],{"type":63,"value":881},"Patterns",{"type":57,"tag":98,"props":883,"children":884},{},[885],{"type":63,"value":886},"Two canonical shapes to orient you; entity-specific examples live in each resource.",{"type":57,"tag":98,"props":888,"children":889},{},[890,892,897],{"type":63,"value":891},"Reach an entity and read fields via ",{"type":57,"tag":104,"props":893,"children":895},{"className":894},[],[896],{"type":63,"value":272},{"type":63,"value":898}," + an inline fragment:",{"type":57,"tag":900,"props":901,"children":905},"pre",{"className":902,"code":903,"language":15,"meta":904,"style":904},"language-graphql shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","query GetEntity($id: ID!) {\n  node(id: $id) {\n    ... on Dataset { name exampleCount }\n  }\n}\n","",[906],{"type":57,"tag":104,"props":907,"children":908},{"__ignoreMap":904},[909,920,929,938,947],{"type":57,"tag":910,"props":911,"children":914},"span",{"class":912,"line":913},"line",1,[915],{"type":57,"tag":910,"props":916,"children":917},{},[918],{"type":63,"value":919},"query GetEntity($id: ID!) {\n",{"type":57,"tag":910,"props":921,"children":923},{"class":912,"line":922},2,[924],{"type":57,"tag":910,"props":925,"children":926},{},[927],{"type":63,"value":928},"  node(id: $id) {\n",{"type":57,"tag":910,"props":930,"children":932},{"class":912,"line":931},3,[933],{"type":57,"tag":910,"props":934,"children":935},{},[936],{"type":63,"value":937},"    ... on Dataset { name exampleCount }\n",{"type":57,"tag":910,"props":939,"children":941},{"class":912,"line":940},4,[942],{"type":57,"tag":910,"props":943,"children":944},{},[945],{"type":63,"value":946},"  }\n",{"type":57,"tag":910,"props":948,"children":950},{"class":912,"line":949},5,[951],{"type":57,"tag":910,"props":952,"children":953},{},[954],{"type":63,"value":955},"}\n",{"type":57,"tag":98,"props":957,"children":958},{},[959],{"type":63,"value":960},"Batch independent project aggregates with aliases in one round trip:",{"type":57,"tag":900,"props":962,"children":964},{"className":902,"code":963,"language":15,"meta":904,"style":904},"query Overview($name: String!, $timeRange: TimeRange) {\n  getProjectByName(name: $name) {\n    traceCount(timeRange: $timeRange)\n    p50: latencyMsQuantile(probability: 0.5, timeRange: $timeRange)\n    p99: latencyMsQuantile(probability: 0.99, timeRange: $timeRange)\n    errorCount: recordCount(timeRange: $timeRange, filterCondition: \"status_code == 'ERROR'\")\n  }\n}\n",[965],{"type":57,"tag":104,"props":966,"children":967},{"__ignoreMap":904},[968,976,984,992,1000,1008,1017,1025],{"type":57,"tag":910,"props":969,"children":970},{"class":912,"line":913},[971],{"type":57,"tag":910,"props":972,"children":973},{},[974],{"type":63,"value":975},"query Overview($name: String!, $timeRange: TimeRange) {\n",{"type":57,"tag":910,"props":977,"children":978},{"class":912,"line":922},[979],{"type":57,"tag":910,"props":980,"children":981},{},[982],{"type":63,"value":983},"  getProjectByName(name: $name) {\n",{"type":57,"tag":910,"props":985,"children":986},{"class":912,"line":931},[987],{"type":57,"tag":910,"props":988,"children":989},{},[990],{"type":63,"value":991},"    traceCount(timeRange: $timeRange)\n",{"type":57,"tag":910,"props":993,"children":994},{"class":912,"line":940},[995],{"type":57,"tag":910,"props":996,"children":997},{},[998],{"type":63,"value":999},"    p50: latencyMsQuantile(probability: 0.5, timeRange: $timeRange)\n",{"type":57,"tag":910,"props":1001,"children":1002},{"class":912,"line":949},[1003],{"type":57,"tag":910,"props":1004,"children":1005},{},[1006],{"type":63,"value":1007},"    p99: latencyMsQuantile(probability: 0.99, timeRange: $timeRange)\n",{"type":57,"tag":910,"props":1009,"children":1011},{"class":912,"line":1010},6,[1012],{"type":57,"tag":910,"props":1013,"children":1014},{},[1015],{"type":63,"value":1016},"    errorCount: recordCount(timeRange: $timeRange, filterCondition: \"status_code == 'ERROR'\")\n",{"type":57,"tag":910,"props":1018,"children":1020},{"class":912,"line":1019},7,[1021],{"type":57,"tag":910,"props":1022,"children":1023},{},[1024],{"type":63,"value":946},{"type":57,"tag":910,"props":1026,"children":1028},{"class":912,"line":1027},8,[1029],{"type":57,"tag":910,"props":1030,"children":1031},{},[1032],{"type":63,"value":955},{"type":57,"tag":58,"props":1034,"children":1036},{"id":1035},"execution-surfaces-internal-mode",[1037],{"type":63,"value":1038},"Execution surfaces (internal mode)",{"type":57,"tag":66,"props":1040,"children":1041},{},[1042],{"type":57,"tag":70,"props":1043,"children":1044},{},[1045,1051,1053,1059,1061,1067,1069,1075,1076,1082,1084,1090],{"type":57,"tag":104,"props":1046,"children":1048},{"className":1047},[],[1049],{"type":63,"value":1050},"phoenix-gql",{"type":63,"value":1052}," (bash): run ",{"type":57,"tag":104,"props":1054,"children":1056},{"className":1055},[],[1057],{"type":63,"value":1058},"phoenix-gql --help",{"type":63,"value":1060}," for flags and current permissions. Use ",{"type":57,"tag":104,"props":1062,"children":1064},{"className":1063},[],[1065],{"type":63,"value":1066},"--data-only",{"type":63,"value":1068}," when piping to ",{"type":57,"tag":104,"props":1070,"children":1072},{"className":1071},[],[1073],{"type":63,"value":1074},"jq",{"type":63,"value":158},{"type":57,"tag":104,"props":1077,"children":1079},{"className":1078},[],[1080],{"type":63,"value":1081},"--output \u003Cfile>",{"type":63,"value":1083}," for large results, ",{"type":57,"tag":104,"props":1085,"children":1087},{"className":1086},[],[1088],{"type":63,"value":1089},"--vars '\u003Cjson>'",{"type":63,"value":1091}," for variables. Mutations are allowed only when runtime permissions say so; the tool reports its permissions on every invocation.",{"type":57,"tag":58,"props":1093,"children":1095},{"id":1094},"external-api-usage-user-facing-mode",[1096],{"type":63,"value":1097},"External API usage (user-facing mode)",{"type":57,"tag":98,"props":1099,"children":1100},{},[1101],{"type":63,"value":1102},"Facts users need to call the API themselves:",{"type":57,"tag":66,"props":1104,"children":1105},{},[1106,1132,1150],{"type":57,"tag":70,"props":1107,"children":1108},{},[1109,1114,1116,1122,1124,1130],{"type":57,"tag":74,"props":1110,"children":1111},{},[1112],{"type":63,"value":1113},"Endpoint",{"type":63,"value":1115},": ",{"type":57,"tag":104,"props":1117,"children":1119},{"className":1118},[],[1120],{"type":63,"value":1121},"POST \u003Cphoenix-host>\u002Fgraphql",{"type":63,"value":1123}," with a JSON body ",{"type":57,"tag":104,"props":1125,"children":1127},{"className":1126},[],[1128],{"type":63,"value":1129},"{ \"query\": \"...\", \"variables\": { ... } }",{"type":63,"value":1131},". A GraphiQL IDE is served on GET at the same path.",{"type":57,"tag":70,"props":1133,"children":1134},{},[1135,1140,1142,1148],{"type":57,"tag":74,"props":1136,"children":1137},{},[1138],{"type":63,"value":1139},"Auth",{"type":63,"value":1141},": send a Phoenix API key as a bearer token: ",{"type":57,"tag":104,"props":1143,"children":1145},{"className":1144},[],[1146],{"type":63,"value":1147},"Authorization: Bearer \u003CAPI_KEY>",{"type":63,"value":1149},". API keys are created in Phoenix settings.",{"type":57,"tag":70,"props":1151,"children":1152},{},[1153,1155,1161,1163,1169,1171,1177],{"type":63,"value":1154},"The GraphQL schema is primarily designed for the Phoenix UI and may change between versions; for stable programmatic access, recommend the REST API (",{"type":57,"tag":104,"props":1156,"children":1158},{"className":1157},[],[1159],{"type":63,"value":1160},"\u002Fv1\u002F...",{"type":63,"value":1162},") and the ",{"type":57,"tag":104,"props":1164,"children":1166},{"className":1165},[],[1167],{"type":63,"value":1168},"arize-phoenix-client",{"type":63,"value":1170}," Python \u002F ",{"type":57,"tag":104,"props":1172,"children":1174},{"className":1173},[],[1175],{"type":63,"value":1176},"@arizeai\u002Fphoenix-client",{"type":63,"value":1178}," TypeScript packages where they cover the need, and GraphQL for everything else.",{"type":57,"tag":98,"props":1180,"children":1181},{},[1182],{"type":63,"value":1183},"curl:",{"type":57,"tag":900,"props":1185,"children":1189},{"className":1186,"code":1187,"language":1188,"meta":904,"style":904},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","curl -s \"$PHOENIX_HOST\u002Fgraphql\" \\\n  -H \"Authorization: Bearer $PHOENIX_API_KEY\" \\\n  -H \"Content-Type: application\u002Fjson\" \\\n  -d '{\"query\": \"query($n: String!) { getProjectByName(name: $n) { traceCount } }\", \"variables\": {\"n\": \"default\"}}'\n","bash",[1190],{"type":57,"tag":104,"props":1191,"children":1192},{"__ignoreMap":904},[1193,1235,1265,1289],{"type":57,"tag":910,"props":1194,"children":1195},{"class":912,"line":913},[1196,1202,1208,1214,1220,1225,1230],{"type":57,"tag":910,"props":1197,"children":1199},{"style":1198},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[1200],{"type":63,"value":1201},"curl",{"type":57,"tag":910,"props":1203,"children":1205},{"style":1204},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[1206],{"type":63,"value":1207}," -s",{"type":57,"tag":910,"props":1209,"children":1211},{"style":1210},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[1212],{"type":63,"value":1213}," \"",{"type":57,"tag":910,"props":1215,"children":1217},{"style":1216},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[1218],{"type":63,"value":1219},"$PHOENIX_HOST",{"type":57,"tag":910,"props":1221,"children":1222},{"style":1204},[1223],{"type":63,"value":1224},"\u002Fgraphql",{"type":57,"tag":910,"props":1226,"children":1227},{"style":1210},[1228],{"type":63,"value":1229},"\"",{"type":57,"tag":910,"props":1231,"children":1232},{"style":1216},[1233],{"type":63,"value":1234}," \\\n",{"type":57,"tag":910,"props":1236,"children":1237},{"class":912,"line":922},[1238,1243,1247,1252,1257,1261],{"type":57,"tag":910,"props":1239,"children":1240},{"style":1204},[1241],{"type":63,"value":1242},"  -H",{"type":57,"tag":910,"props":1244,"children":1245},{"style":1210},[1246],{"type":63,"value":1213},{"type":57,"tag":910,"props":1248,"children":1249},{"style":1204},[1250],{"type":63,"value":1251},"Authorization: Bearer ",{"type":57,"tag":910,"props":1253,"children":1254},{"style":1216},[1255],{"type":63,"value":1256},"$PHOENIX_API_KEY",{"type":57,"tag":910,"props":1258,"children":1259},{"style":1210},[1260],{"type":63,"value":1229},{"type":57,"tag":910,"props":1262,"children":1263},{"style":1216},[1264],{"type":63,"value":1234},{"type":57,"tag":910,"props":1266,"children":1267},{"class":912,"line":931},[1268,1272,1276,1281,1285],{"type":57,"tag":910,"props":1269,"children":1270},{"style":1204},[1271],{"type":63,"value":1242},{"type":57,"tag":910,"props":1273,"children":1274},{"style":1210},[1275],{"type":63,"value":1213},{"type":57,"tag":910,"props":1277,"children":1278},{"style":1204},[1279],{"type":63,"value":1280},"Content-Type: application\u002Fjson",{"type":57,"tag":910,"props":1282,"children":1283},{"style":1210},[1284],{"type":63,"value":1229},{"type":57,"tag":910,"props":1286,"children":1287},{"style":1216},[1288],{"type":63,"value":1234},{"type":57,"tag":910,"props":1290,"children":1291},{"class":912,"line":940},[1292,1297,1302,1307],{"type":57,"tag":910,"props":1293,"children":1294},{"style":1204},[1295],{"type":63,"value":1296},"  -d",{"type":57,"tag":910,"props":1298,"children":1299},{"style":1210},[1300],{"type":63,"value":1301}," '",{"type":57,"tag":910,"props":1303,"children":1304},{"style":1204},[1305],{"type":63,"value":1306},"{\"query\": \"query($n: String!) { getProjectByName(name: $n) { traceCount } }\", \"variables\": {\"n\": \"default\"}}",{"type":57,"tag":910,"props":1308,"children":1309},{"style":1210},[1310],{"type":63,"value":1311},"'\n",{"type":57,"tag":98,"props":1313,"children":1314},{},[1315],{"type":63,"value":1316},"Python:",{"type":57,"tag":900,"props":1318,"children":1322},{"className":1319,"code":1320,"language":1321,"meta":904,"style":904},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","import httpx\n\nresp = httpx.post(\n    f\"{host}\u002Fgraphql\",\n    headers={\"Authorization\": f\"Bearer {api_key}\"},\n    json={\"query\": query, \"variables\": variables},\n)\nresp.raise_for_status()\ndata = resp.json()[\"data\"]\n","python",[1323],{"type":57,"tag":104,"props":1324,"children":1325},{"__ignoreMap":904},[1326,1334,1343,1351,1359,1367,1375,1383,1391],{"type":57,"tag":910,"props":1327,"children":1328},{"class":912,"line":913},[1329],{"type":57,"tag":910,"props":1330,"children":1331},{},[1332],{"type":63,"value":1333},"import httpx\n",{"type":57,"tag":910,"props":1335,"children":1336},{"class":912,"line":922},[1337],{"type":57,"tag":910,"props":1338,"children":1340},{"emptyLinePlaceholder":1339},true,[1341],{"type":63,"value":1342},"\n",{"type":57,"tag":910,"props":1344,"children":1345},{"class":912,"line":931},[1346],{"type":57,"tag":910,"props":1347,"children":1348},{},[1349],{"type":63,"value":1350},"resp = httpx.post(\n",{"type":57,"tag":910,"props":1352,"children":1353},{"class":912,"line":940},[1354],{"type":57,"tag":910,"props":1355,"children":1356},{},[1357],{"type":63,"value":1358},"    f\"{host}\u002Fgraphql\",\n",{"type":57,"tag":910,"props":1360,"children":1361},{"class":912,"line":949},[1362],{"type":57,"tag":910,"props":1363,"children":1364},{},[1365],{"type":63,"value":1366},"    headers={\"Authorization\": f\"Bearer {api_key}\"},\n",{"type":57,"tag":910,"props":1368,"children":1369},{"class":912,"line":1010},[1370],{"type":57,"tag":910,"props":1371,"children":1372},{},[1373],{"type":63,"value":1374},"    json={\"query\": query, \"variables\": variables},\n",{"type":57,"tag":910,"props":1376,"children":1377},{"class":912,"line":1019},[1378],{"type":57,"tag":910,"props":1379,"children":1380},{},[1381],{"type":63,"value":1382},")\n",{"type":57,"tag":910,"props":1384,"children":1385},{"class":912,"line":1027},[1386],{"type":57,"tag":910,"props":1387,"children":1388},{},[1389],{"type":63,"value":1390},"resp.raise_for_status()\n",{"type":57,"tag":910,"props":1392,"children":1394},{"class":912,"line":1393},9,[1395],{"type":57,"tag":910,"props":1396,"children":1397},{},[1398],{"type":63,"value":1399},"data = resp.json()[\"data\"]\n",{"type":57,"tag":98,"props":1401,"children":1402},{},[1403,1405,1410,1412,1418,1420,1425],{"type":63,"value":1404},"When handing users a query, include: the full operation with variable definitions, an example variables payload, and a note on paginating via ",{"type":57,"tag":104,"props":1406,"children":1408},{"className":1407},[],[1409],{"type":63,"value":488},{"type":63,"value":1411}," → pass ",{"type":57,"tag":104,"props":1413,"children":1415},{"className":1414},[],[1416],{"type":63,"value":1417},"endCursor",{"type":63,"value":1419}," as ",{"type":57,"tag":104,"props":1421,"children":1423},{"className":1422},[],[1424],{"type":63,"value":472},{"type":63,"value":708},{"type":57,"tag":1427,"props":1428,"children":1429},"style",{},[1430],{"type":63,"value":1431},"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":1433,"total":1591},[1434,1452,1462,1474,1486,1496,1502,1512,1523,1541,1559,1569],{"slug":1435,"name":1435,"fn":1436,"description":1437,"org":1438,"tags":1439,"stars":23,"repoUrl":24,"updatedAt":1451},"annotate-spans","annotate LLM spans and traces","Write effective, consistent annotations on LLM\u002Fagent spans and traces, and coach the user on annotation practice. Load this whenever you are about to record structured feedback with the `batch_span_annotate` tool, or when the user asks how to annotate, label, score, or review spans\u002Ftraces, build a failure taxonomy, or set up human\u002FLLM review. Do NOT load for: pure analysis with no intent to save feedback (use debug-trace), latency or cost statistics, or prompt authoring (use playground).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1440,1442,1445,1448],{"name":1441,"slug":35,"type":16},"Evals",{"name":1443,"slug":1444,"type":16},"LLM","llm",{"name":1446,"slug":1447,"type":16},"Observability","observability",{"name":1449,"slug":1450,"type":16},"Tracing","tracing","2026-07-12T08:08:14.140984",{"slug":34,"name":34,"fn":1453,"description":1454,"org":1455,"tags":1456,"stars":23,"repoUrl":24,"updatedAt":1461},"reason about Phoenix dataset structure","Understand what a Phoenix dataset is and reason well about its examples, outputs, splits, and how it feeds evaluators and experiments. Load this whenever a dataset is in view or the user asks what a dataset is, how splits work, what an output \"means\", or how datasets relate to experiments and evals. This skill governs the judgment; any tool descriptions govern the mechanics.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1457,1458,1460],{"name":18,"slug":19,"type":16},{"name":1459,"slug":34,"type":16},"Datasets",{"name":1441,"slug":35,"type":16},"2026-07-12T08:08:21.695457",{"slug":1463,"name":1463,"fn":1464,"description":1465,"org":1466,"tags":1467,"stars":23,"repoUrl":24,"updatedAt":1473},"debug-trace","diagnose failures using trace investigation","Diagnose failure modes by systematically investigating traces. Trigger when the user explicitly asks for cross-trace diagnosis: \"what's going wrong?\", \"were there errors?\", \"debug this\", \"where is my agent struggling?\". Do NOT trigger on: (1) advice questions (\"what should I do?\"), (2) statistical questions (\"what's the average latency?\"), (3) summarize requests, (4) trace filtering (\"show me traces with errors\"), (5) vague questions (\"is there a problem?\"), (6) unrelated requests.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1468,1471,1472],{"name":1469,"slug":1470,"type":16},"Debugging","debugging",{"name":1446,"slug":1447,"type":16},{"name":1449,"slug":1450,"type":16},"2026-07-12T08:08:10.44243",{"slug":1475,"name":1475,"fn":1476,"description":1477,"org":1478,"tags":1479,"stars":23,"repoUrl":24,"updatedAt":1485},"evaluators","author and refine Phoenix evaluators","Author or refine a Phoenix evaluator — code or LLM-as-a-judge — that scores a run's output. Trigger when the user wants to create a new evaluator, improve an existing one's logic or rubric, choose labels, or decide what to measure on a dataset or experiment. Do NOT trigger on: (1) manual prompt drafting (use `playground`), (2) running or comparing experiments themselves (use `experiments`), (3) cross-trace failure diagnosis with no evaluator in scope (use `debug-trace`).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1480,1481,1482],{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1483,"slug":1484,"type":16},"Testing","testing","2026-07-31T05:58:09.13624",{"slug":411,"name":411,"fn":1487,"description":1488,"org":1489,"tags":1490,"stars":23,"repoUrl":24,"updatedAt":1495},"run and compare dataset-backed experiments","Run, read, and compare dataset-backed experiments to find evidence that a prompt or pipeline is improving. Trigger when the user wants to iterate over a dataset with experiments, compare experiment runs, read experiment quality\u002Flatency\u002Fcost, or decide whether a change actually helped. Running a prompt over a dataset is implicitly an experiment — load this skill when dataset-backed work begins, before authoring evaluators for the experiment and before starting the recorded run, not only when reading results. Do NOT trigger on: (1) manual prompt drafting with no dataset-backed evaluation in scope (use `playground`), (2) authoring or refining an evaluator's logic or rubric (use `evaluators`), (3) cross-trace failure diagnosis with no experiment in scope (use `debug-trace`).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1491,1492,1493,1494],{"name":1459,"slug":34,"type":16},{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1483,"slug":1484,"type":16},"2026-07-12T08:08:11.691477",{"slug":4,"name":4,"fn":5,"description":6,"org":1497,"tags":1498,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1499,1500,1501],{"name":21,"slug":22,"type":16},{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"slug":1503,"name":1503,"fn":1504,"description":1505,"org":1506,"tags":1507,"stars":23,"repoUrl":24,"updatedAt":1511},"playground","author and iterate on prompts in Phoenix","Author, edit, or iterate on prompts in the Phoenix prompt playground, including running experiments over a dataset. Load before any playground tool call, including single-shot prompt rewrites.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1508,1509,1510],{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1483,"slug":1484,"type":16},"2026-07-12T08:08:12.920792",{"slug":1513,"name":1513,"fn":1514,"description":1515,"org":1516,"tags":1517,"stars":23,"repoUrl":24,"updatedAt":1522},"span-coding","analyze and code Phoenix spans","Open-code Phoenix spans with PXI-owned notes, recover those notes for axial coding, and promote stable categories into structured annotations. Load this when analyzing spans to discover failure patterns before a taxonomy exists.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1518,1519,1520,1521],{"name":1469,"slug":1470,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1446,"slug":1447,"type":16},{"name":1449,"slug":1450,"type":16},"2026-07-12T08:08:19.597239",{"slug":1524,"name":1524,"fn":1525,"description":1526,"org":1527,"tags":1528,"stars":1538,"repoUrl":1539,"updatedAt":1540},"arize-admin","manage Arize enterprise user access","Manages Arize users, organizations, spaces, projects, roles, role bindings, resource restrictions, and API keys via the ax CLI. Use for enterprise admin workflows: inviting and offboarding users, onboarding new teams, creating custom roles for SAML\u002FSSO mappings, assigning roles to users, restricting project-level access, and managing service keys for multi-tenant architectures. Covers ax users, ax organizations, ax spaces, ax projects, ax roles, ax role-bindings, and ax api-keys.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1529,1532,1535],{"name":1530,"slug":1531,"type":16},"CLI","cli",{"name":1533,"slug":1534,"type":16},"Operations","operations",{"name":1536,"slug":1537,"type":16},"Permissions","permissions",38,"https:\u002F\u002Fgithub.com\u002FArize-ai\u002Farize-skills","2026-07-22T05:37:21.991338",{"slug":1542,"name":1542,"fn":1543,"description":1544,"org":1545,"tags":1546,"stars":1538,"repoUrl":1539,"updatedAt":1558},"arize-ai-provider-integration","manage Arize AI provider integrations","Creates, reads, updates, and deletes Arize AI integrations that store LLM provider credentials used by evaluators and other Arize features. Supports any LLM provider (e.g. OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, Vertex AI, Gemini, NVIDIA NIM). Use when the user mentions AI integration, LLM provider credentials, create integration, list integrations, update credentials, delete integration, or connecting an LLM provider to Arize.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1547,1549,1552,1555,1556],{"name":1548,"slug":33,"type":16},"Anthropic",{"name":1550,"slug":1551,"type":16},"Azure","azure",{"name":1553,"slug":1554,"type":16},"Integrations","integrations",{"name":1443,"slug":1444,"type":16},{"name":1557,"slug":42,"type":16},"OpenAI","2026-07-22T05:37:23.90468",{"slug":1560,"name":1560,"fn":1561,"description":1562,"org":1563,"tags":1564,"stars":1538,"repoUrl":1539,"updatedAt":1568},"arize-annotation","manage Arize annotation workflows","Creates and manages annotation configs (categorical, continuous, freeform label schemas) and annotation queues (human review workflows) on Arize. Applies human annotations to project spans via the Python SDK. Use when the user mentions annotation config, annotation queue, label schema, human feedback, bulk annotate spans, update_annotations, labeling queue, annotate record, or human review.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1565,1566,1567],{"name":18,"slug":19,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1446,"slug":1447,"type":16},"2026-07-22T05:37:19.010776",{"slug":1570,"name":1570,"fn":1571,"description":1572,"org":1573,"tags":1574,"stars":1538,"repoUrl":1539,"updatedAt":1590},"arize-compliance-audit","audit AI agents for regulatory compliance","INVOKE THIS SKILL when auditing an AI agent or LLM app for regulatory compliance. Covers EU AI Act, GPAI Code of Practice, GDPR, NIST AI RMF, Colorado AI Act, HIPAA, and ISO 42001. Scans the codebase for compliance gaps, cross-references Arize instrumentation for audit trail coverage, and produces an actionable remediation checklist tailored to the selected frameworks.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1575,1578,1581,1584,1587],{"name":1576,"slug":1577,"type":16},"Audit","audit",{"name":1579,"slug":1580,"type":16},"Compliance","compliance",{"name":1582,"slug":1583,"type":16},"GDPR","gdpr",{"name":1585,"slug":1586,"type":16},"Legal","legal",{"name":1588,"slug":1589,"type":16},"Security","security","2026-07-19T05:39:42.632738",23,{"items":1593,"total":1027},[1594,1601,1607,1613,1619,1626,1632],{"slug":1435,"name":1435,"fn":1436,"description":1437,"org":1595,"tags":1596,"stars":23,"repoUrl":24,"updatedAt":1451},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1597,1598,1599,1600],{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1446,"slug":1447,"type":16},{"name":1449,"slug":1450,"type":16},{"slug":34,"name":34,"fn":1453,"description":1454,"org":1602,"tags":1603,"stars":23,"repoUrl":24,"updatedAt":1461},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1604,1605,1606],{"name":18,"slug":19,"type":16},{"name":1459,"slug":34,"type":16},{"name":1441,"slug":35,"type":16},{"slug":1463,"name":1463,"fn":1464,"description":1465,"org":1608,"tags":1609,"stars":23,"repoUrl":24,"updatedAt":1473},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1610,1611,1612],{"name":1469,"slug":1470,"type":16},{"name":1446,"slug":1447,"type":16},{"name":1449,"slug":1450,"type":16},{"slug":1475,"name":1475,"fn":1476,"description":1477,"org":1614,"tags":1615,"stars":23,"repoUrl":24,"updatedAt":1485},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1616,1617,1618],{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1483,"slug":1484,"type":16},{"slug":411,"name":411,"fn":1487,"description":1488,"org":1620,"tags":1621,"stars":23,"repoUrl":24,"updatedAt":1495},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1622,1623,1624,1625],{"name":1459,"slug":34,"type":16},{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1483,"slug":1484,"type":16},{"slug":4,"name":4,"fn":5,"description":6,"org":1627,"tags":1628,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1629,1630,1631],{"name":21,"slug":22,"type":16},{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"slug":1503,"name":1503,"fn":1504,"description":1505,"org":1633,"tags":1634,"stars":23,"repoUrl":24,"updatedAt":1511},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1635,1636,1637],{"name":1441,"slug":35,"type":16},{"name":1443,"slug":1444,"type":16},{"name":1483,"slug":1484,"type":16}]