[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-mlflow-instrumenting-with-mlflow-tracing":3,"mdc-xepnr6-key":42,"related-org-mlflow-instrumenting-with-mlflow-tracing":902,"related-repo-mlflow-instrumenting-with-mlflow-tracing":1057},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":32,"repoUrl":33,"updatedAt":34,"license":35,"forks":36,"topics":37,"repo":38,"sourceUrl":40,"mdContent":41},"instrumenting-with-mlflow-tracing","instrument Python and TypeScript with MLflow Tracing","Instruments Python and TypeScript code with MLflow Tracing for observability. Must be loaded when setting up tracing as part of any workflow including agent evaluation. Triggers on adding tracing, instrumenting agents\u002FLLM apps, getting started with MLflow tracing, tracing specific frameworks (LangGraph, LangChain, OpenAI, Gemini, DSPy, CrewAI, AutoGen), or when another skill references tracing setup. Examples - \"How do I add tracing?\", \"Instrument my agent\", \"Trace my LangChain app\", \"Set up tracing for evaluation\"",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"mlflow","MLflow","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fmlflow.png",[12,16,19,20,23,26,29],{"name":13,"slug":14,"type":15},"Tracing","tracing","tag",{"name":17,"slug":18,"type":15},"Observability","observability",{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},"TypeScript","typescript",{"name":24,"slug":25,"type":15},"Evals","evals",{"name":27,"slug":28,"type":15},"Python","python",{"name":30,"slug":31,"type":15},"Agents","agents",60,"https:\u002F\u002Fgithub.com\u002Fmlflow\u002Fskills","2026-07-20T05:58:52.968218",null,19,[],{"repoUrl":33,"stars":32,"forks":36,"topics":39,"description":35},[],"https:\u002F\u002Fgithub.com\u002Fmlflow\u002Fskills\u002Ftree\u002FHEAD\u002Finstrumenting-with-mlflow-tracing","---\nname: instrumenting-with-mlflow-tracing\ndescription: Instruments Python and TypeScript code with MLflow Tracing for observability. Must be loaded when setting up tracing as part of any workflow including agent evaluation. Triggers on adding tracing, instrumenting agents\u002FLLM apps, getting started with MLflow tracing, tracing specific frameworks (LangGraph, LangChain, OpenAI, Gemini, DSPy, CrewAI, AutoGen), or when another skill references tracing setup. Examples - \"How do I add tracing?\", \"Instrument my agent\", \"Trace my LangChain app\", \"Set up tracing for evaluation\"\n---\n\n# MLflow Tracing Instrumentation Guide\n\n## Language-Specific Guides\n\nBased on the user's project, load the appropriate guide:\n\n- **Python projects**: Read `references\u002Fpython.md`\n- **TypeScript\u002FJavaScript projects**: Read `references\u002Ftypescript.md`\n\nIf unclear, check for `package.json` (TypeScript) or `requirements.txt`\u002F`pyproject.toml` (Python) in the project.\n\n---\n\n## What to Trace\n\n**Trace these operations** (high debugging\u002Fobservability value):\n\n| Operation Type | Examples | Why Trace |\n|---------------|----------|-----------|\n| **Root operations** | Main entry points, top-level pipelines, workflow steps | End-to-end latency, input\u002Foutput logging |\n| **LLM calls** | Chat completions, embeddings | Token usage, latency, prompt\u002Fresponse inspection |\n| **Retrieval** | Vector DB queries, document fetches, search | Relevance debugging, retrieval quality |\n| **Tool\u002Ffunction calls** | API calls, database queries, web search | External dependency monitoring, error tracking |\n| **Agent decisions** | Routing, planning, tool selection | Understand agent reasoning and choices |\n| **External services** | HTTP APIs, file I\u002FO, message queues | Dependency failures, timeout tracking |\n\n**Skip tracing these** (too granular, adds noise):\n\n- Simple data transformations (dict\u002Flist manipulation)\n- String formatting, parsing, validation\n- Configuration loading, environment setup\n- Logging or metric emission\n- Pure utility functions (math, sorting, filtering)\n\n**Rule of thumb**: Trace operations that are important for debugging and identifying issues in your application.\n\n---\n\n## Verification\n\nAfter instrumenting the code, **always verify that tracing is working**.\n\n> **Planning to evaluate your agent?** Tracing must be working before you run `agent-evaluation`. Complete verification below first.\n\n\n1. **Run the instrumented code** — execute the application or agent so that at least one traced operation fires\n2. **Confirm traces are logged** — use `mlflow.search_traces()` or `MlflowClient().search_traces()` to check that traces appear in the experiment:\n\n```python\nimport mlflow\n\ntraces = mlflow.search_traces(experiment_ids=[\"\u003Cexperiment_id>\"])\nprint(f\"Found {len(traces)} trace(s)\")\nassert len(traces) > 0, \"No traces were logged — check tracking URI and experiment settings\"\n```\n\n3. **Verify spans were captured** — confirm the trace contains the expected spans, not just an empty shell:\n\n```python\ntrace = traces.iloc[0]\nspans = mlflow.get_trace(trace.trace_id).data.spans\nprint(f\"Trace has {len(spans)} span(s)\")\nfor span in spans:\n    print(f\"  - {span.name} ({span.span_type})\")\n```\n\n4. **Report the result** — tell the user how many traces and spans were found and confirm tracing is working\n\n### If no traces appear\n\nCheck these in order:\n\n- **Tracking URI not set** — is `mlflow.set_tracking_uri(...)` called before the agent run? Without this, traces go to a local `.\u002Fmlruns` directory instead of the configured server.\n- **Autolog warnings** — did `mlflow.autolog()` or framework-specific `mlflow.\u003Cframework>.autolog()` raise any warnings during setup? Check stderr for patching failures.\n- **Wrong experiment ID** — verify the experiment ID passed to `search_traces()` matches the experiment active when the code ran (`mlflow.get_experiment_by_name(...)` to confirm).\n- **Network\u002Fauth issues** — can the process reach the tracking server? Check for connection errors or 401\u002F403 responses in logs.\n\nFor automated validation, use `agent-evaluation\u002Fscripts\u002Fvalidate_tracing_runtime.py`.\n\n---\n\n## Feedback Collection\n\nLog user feedback on traces for evaluation, debugging, and fine-tuning. Essential for identifying quality issues in production.\n\nSee `references\u002Ffeedback-collection.md` for:\n- Recording user ratings and comments with `mlflow.log_feedback()`\n- Capturing trace IDs to return to clients\n- LLM-as-judge automated evaluation\n\n---\n\n## Reference Documentation\n\n### Production Deployment\n\nSee `references\u002Fproduction.md` for:\n- Environment variable configuration\n- Async logging for low-latency applications\n- Sampling configuration (MLFLOW_TRACE_SAMPLING_RATIO)\n- Lightweight SDK (`mlflow-tracing`)\n- Docker\u002FKubernetes deployment\n\n### Advanced Patterns\n\nSee `references\u002Fadvanced-patterns.md` for:\n- Async function tracing\n- Multi-threading with context propagation\n- PII redaction with span processors\n\n### Distributed Tracing\n\nSee `references\u002Fdistributed-tracing.md` for:\n- Propagating trace context across services\n- Client\u002Fserver header APIs\n\n### Databricks (Unity Catalog storage)\n\nSee `references\u002Fdatabricks.md` for storing traces in Unity Catalog Delta tables by binding an experiment to a `UnityCatalog` trace location (catalog, schema, table prefix).\n",{"data":43,"body":44},{"name":4,"description":6},{"type":45,"children":46},"root",[47,56,63,69,107,136,140,146,156,315,325,353,363,366,372,384,406,446,504,517,564,577,584,589,680,692,695,701,706,719,743,746,752,758,769,805,811,822,840,846,857,870,876,896],{"type":48,"tag":49,"props":50,"children":52},"element","h1",{"id":51},"mlflow-tracing-instrumentation-guide",[53],{"type":54,"value":55},"text","MLflow Tracing Instrumentation Guide",{"type":48,"tag":57,"props":58,"children":60},"h2",{"id":59},"language-specific-guides",[61],{"type":54,"value":62},"Language-Specific Guides",{"type":48,"tag":64,"props":65,"children":66},"p",{},[67],{"type":54,"value":68},"Based on the user's project, load the appropriate guide:",{"type":48,"tag":70,"props":71,"children":72},"ul",{},[73,92],{"type":48,"tag":74,"props":75,"children":76},"li",{},[77,83,85],{"type":48,"tag":78,"props":79,"children":80},"strong",{},[81],{"type":54,"value":82},"Python projects",{"type":54,"value":84},": Read ",{"type":48,"tag":86,"props":87,"children":89},"code",{"className":88},[],[90],{"type":54,"value":91},"references\u002Fpython.md",{"type":48,"tag":74,"props":93,"children":94},{},[95,100,101],{"type":48,"tag":78,"props":96,"children":97},{},[98],{"type":54,"value":99},"TypeScript\u002FJavaScript projects",{"type":54,"value":84},{"type":48,"tag":86,"props":102,"children":104},{"className":103},[],[105],{"type":54,"value":106},"references\u002Ftypescript.md",{"type":48,"tag":64,"props":108,"children":109},{},[110,112,118,120,126,128,134],{"type":54,"value":111},"If unclear, check for ",{"type":48,"tag":86,"props":113,"children":115},{"className":114},[],[116],{"type":54,"value":117},"package.json",{"type":54,"value":119}," (TypeScript) or ",{"type":48,"tag":86,"props":121,"children":123},{"className":122},[],[124],{"type":54,"value":125},"requirements.txt",{"type":54,"value":127},"\u002F",{"type":48,"tag":86,"props":129,"children":131},{"className":130},[],[132],{"type":54,"value":133},"pyproject.toml",{"type":54,"value":135}," (Python) in the project.",{"type":48,"tag":137,"props":138,"children":139},"hr",{},[],{"type":48,"tag":57,"props":141,"children":143},{"id":142},"what-to-trace",[144],{"type":54,"value":145},"What to Trace",{"type":48,"tag":64,"props":147,"children":148},{},[149,154],{"type":48,"tag":78,"props":150,"children":151},{},[152],{"type":54,"value":153},"Trace these operations",{"type":54,"value":155}," (high debugging\u002Fobservability value):",{"type":48,"tag":157,"props":158,"children":159},"table",{},[160,184],{"type":48,"tag":161,"props":162,"children":163},"thead",{},[164],{"type":48,"tag":165,"props":166,"children":167},"tr",{},[168,174,179],{"type":48,"tag":169,"props":170,"children":171},"th",{},[172],{"type":54,"value":173},"Operation Type",{"type":48,"tag":169,"props":175,"children":176},{},[177],{"type":54,"value":178},"Examples",{"type":48,"tag":169,"props":180,"children":181},{},[182],{"type":54,"value":183},"Why Trace",{"type":48,"tag":185,"props":186,"children":187},"tbody",{},[188,210,231,252,273,294],{"type":48,"tag":165,"props":189,"children":190},{},[191,200,205],{"type":48,"tag":192,"props":193,"children":194},"td",{},[195],{"type":48,"tag":78,"props":196,"children":197},{},[198],{"type":54,"value":199},"Root operations",{"type":48,"tag":192,"props":201,"children":202},{},[203],{"type":54,"value":204},"Main entry points, top-level pipelines, workflow steps",{"type":48,"tag":192,"props":206,"children":207},{},[208],{"type":54,"value":209},"End-to-end latency, input\u002Foutput logging",{"type":48,"tag":165,"props":211,"children":212},{},[213,221,226],{"type":48,"tag":192,"props":214,"children":215},{},[216],{"type":48,"tag":78,"props":217,"children":218},{},[219],{"type":54,"value":220},"LLM calls",{"type":48,"tag":192,"props":222,"children":223},{},[224],{"type":54,"value":225},"Chat completions, embeddings",{"type":48,"tag":192,"props":227,"children":228},{},[229],{"type":54,"value":230},"Token usage, latency, prompt\u002Fresponse inspection",{"type":48,"tag":165,"props":232,"children":233},{},[234,242,247],{"type":48,"tag":192,"props":235,"children":236},{},[237],{"type":48,"tag":78,"props":238,"children":239},{},[240],{"type":54,"value":241},"Retrieval",{"type":48,"tag":192,"props":243,"children":244},{},[245],{"type":54,"value":246},"Vector DB queries, document fetches, search",{"type":48,"tag":192,"props":248,"children":249},{},[250],{"type":54,"value":251},"Relevance debugging, retrieval quality",{"type":48,"tag":165,"props":253,"children":254},{},[255,263,268],{"type":48,"tag":192,"props":256,"children":257},{},[258],{"type":48,"tag":78,"props":259,"children":260},{},[261],{"type":54,"value":262},"Tool\u002Ffunction calls",{"type":48,"tag":192,"props":264,"children":265},{},[266],{"type":54,"value":267},"API calls, database queries, web search",{"type":48,"tag":192,"props":269,"children":270},{},[271],{"type":54,"value":272},"External dependency monitoring, error tracking",{"type":48,"tag":165,"props":274,"children":275},{},[276,284,289],{"type":48,"tag":192,"props":277,"children":278},{},[279],{"type":48,"tag":78,"props":280,"children":281},{},[282],{"type":54,"value":283},"Agent decisions",{"type":48,"tag":192,"props":285,"children":286},{},[287],{"type":54,"value":288},"Routing, planning, tool selection",{"type":48,"tag":192,"props":290,"children":291},{},[292],{"type":54,"value":293},"Understand agent reasoning and choices",{"type":48,"tag":165,"props":295,"children":296},{},[297,305,310],{"type":48,"tag":192,"props":298,"children":299},{},[300],{"type":48,"tag":78,"props":301,"children":302},{},[303],{"type":54,"value":304},"External services",{"type":48,"tag":192,"props":306,"children":307},{},[308],{"type":54,"value":309},"HTTP APIs, file I\u002FO, message queues",{"type":48,"tag":192,"props":311,"children":312},{},[313],{"type":54,"value":314},"Dependency failures, timeout tracking",{"type":48,"tag":64,"props":316,"children":317},{},[318,323],{"type":48,"tag":78,"props":319,"children":320},{},[321],{"type":54,"value":322},"Skip tracing these",{"type":54,"value":324}," (too granular, adds noise):",{"type":48,"tag":70,"props":326,"children":327},{},[328,333,338,343,348],{"type":48,"tag":74,"props":329,"children":330},{},[331],{"type":54,"value":332},"Simple data transformations (dict\u002Flist manipulation)",{"type":48,"tag":74,"props":334,"children":335},{},[336],{"type":54,"value":337},"String formatting, parsing, validation",{"type":48,"tag":74,"props":339,"children":340},{},[341],{"type":54,"value":342},"Configuration loading, environment setup",{"type":48,"tag":74,"props":344,"children":345},{},[346],{"type":54,"value":347},"Logging or metric emission",{"type":48,"tag":74,"props":349,"children":350},{},[351],{"type":54,"value":352},"Pure utility functions (math, sorting, filtering)",{"type":48,"tag":64,"props":354,"children":355},{},[356,361],{"type":48,"tag":78,"props":357,"children":358},{},[359],{"type":54,"value":360},"Rule of thumb",{"type":54,"value":362},": Trace operations that are important for debugging and identifying issues in your application.",{"type":48,"tag":137,"props":364,"children":365},{},[],{"type":48,"tag":57,"props":367,"children":369},{"id":368},"verification",[370],{"type":54,"value":371},"Verification",{"type":48,"tag":64,"props":373,"children":374},{},[375,377,382],{"type":54,"value":376},"After instrumenting the code, ",{"type":48,"tag":78,"props":378,"children":379},{},[380],{"type":54,"value":381},"always verify that tracing is working",{"type":54,"value":383},".",{"type":48,"tag":385,"props":386,"children":387},"blockquote",{},[388],{"type":48,"tag":64,"props":389,"children":390},{},[391,396,398,404],{"type":48,"tag":78,"props":392,"children":393},{},[394],{"type":54,"value":395},"Planning to evaluate your agent?",{"type":54,"value":397}," Tracing must be working before you run ",{"type":48,"tag":86,"props":399,"children":401},{"className":400},[],[402],{"type":54,"value":403},"agent-evaluation",{"type":54,"value":405},". Complete verification below first.",{"type":48,"tag":407,"props":408,"children":409},"ol",{},[410,420],{"type":48,"tag":74,"props":411,"children":412},{},[413,418],{"type":48,"tag":78,"props":414,"children":415},{},[416],{"type":54,"value":417},"Run the instrumented code",{"type":54,"value":419}," — execute the application or agent so that at least one traced operation fires",{"type":48,"tag":74,"props":421,"children":422},{},[423,428,430,436,438,444],{"type":48,"tag":78,"props":424,"children":425},{},[426],{"type":54,"value":427},"Confirm traces are logged",{"type":54,"value":429}," — use ",{"type":48,"tag":86,"props":431,"children":433},{"className":432},[],[434],{"type":54,"value":435},"mlflow.search_traces()",{"type":54,"value":437}," or ",{"type":48,"tag":86,"props":439,"children":441},{"className":440},[],[442],{"type":54,"value":443},"MlflowClient().search_traces()",{"type":54,"value":445}," to check that traces appear in the experiment:",{"type":48,"tag":447,"props":448,"children":452},"pre",{"className":449,"code":450,"language":28,"meta":451,"style":451},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","import mlflow\n\ntraces = mlflow.search_traces(experiment_ids=[\"\u003Cexperiment_id>\"])\nprint(f\"Found {len(traces)} trace(s)\")\nassert len(traces) > 0, \"No traces were logged — check tracking URI and experiment settings\"\n","",[453],{"type":48,"tag":86,"props":454,"children":455},{"__ignoreMap":451},[456,467,477,486,495],{"type":48,"tag":457,"props":458,"children":461},"span",{"class":459,"line":460},"line",1,[462],{"type":48,"tag":457,"props":463,"children":464},{},[465],{"type":54,"value":466},"import mlflow\n",{"type":48,"tag":457,"props":468,"children":470},{"class":459,"line":469},2,[471],{"type":48,"tag":457,"props":472,"children":474},{"emptyLinePlaceholder":473},true,[475],{"type":54,"value":476},"\n",{"type":48,"tag":457,"props":478,"children":480},{"class":459,"line":479},3,[481],{"type":48,"tag":457,"props":482,"children":483},{},[484],{"type":54,"value":485},"traces = mlflow.search_traces(experiment_ids=[\"\u003Cexperiment_id>\"])\n",{"type":48,"tag":457,"props":487,"children":489},{"class":459,"line":488},4,[490],{"type":48,"tag":457,"props":491,"children":492},{},[493],{"type":54,"value":494},"print(f\"Found {len(traces)} trace(s)\")\n",{"type":48,"tag":457,"props":496,"children":498},{"class":459,"line":497},5,[499],{"type":48,"tag":457,"props":500,"children":501},{},[502],{"type":54,"value":503},"assert len(traces) > 0, \"No traces were logged — check tracking URI and experiment settings\"\n",{"type":48,"tag":407,"props":505,"children":506},{"start":479},[507],{"type":48,"tag":74,"props":508,"children":509},{},[510,515],{"type":48,"tag":78,"props":511,"children":512},{},[513],{"type":54,"value":514},"Verify spans were captured",{"type":54,"value":516}," — confirm the trace contains the expected spans, not just an empty shell:",{"type":48,"tag":447,"props":518,"children":520},{"className":449,"code":519,"language":28,"meta":451,"style":451},"trace = traces.iloc[0]\nspans = mlflow.get_trace(trace.trace_id).data.spans\nprint(f\"Trace has {len(spans)} span(s)\")\nfor span in spans:\n    print(f\"  - {span.name} ({span.span_type})\")\n",[521],{"type":48,"tag":86,"props":522,"children":523},{"__ignoreMap":451},[524,532,540,548,556],{"type":48,"tag":457,"props":525,"children":526},{"class":459,"line":460},[527],{"type":48,"tag":457,"props":528,"children":529},{},[530],{"type":54,"value":531},"trace = traces.iloc[0]\n",{"type":48,"tag":457,"props":533,"children":534},{"class":459,"line":469},[535],{"type":48,"tag":457,"props":536,"children":537},{},[538],{"type":54,"value":539},"spans = mlflow.get_trace(trace.trace_id).data.spans\n",{"type":48,"tag":457,"props":541,"children":542},{"class":459,"line":479},[543],{"type":48,"tag":457,"props":544,"children":545},{},[546],{"type":54,"value":547},"print(f\"Trace has {len(spans)} span(s)\")\n",{"type":48,"tag":457,"props":549,"children":550},{"class":459,"line":488},[551],{"type":48,"tag":457,"props":552,"children":553},{},[554],{"type":54,"value":555},"for span in spans:\n",{"type":48,"tag":457,"props":557,"children":558},{"class":459,"line":497},[559],{"type":48,"tag":457,"props":560,"children":561},{},[562],{"type":54,"value":563},"    print(f\"  - {span.name} ({span.span_type})\")\n",{"type":48,"tag":407,"props":565,"children":566},{"start":488},[567],{"type":48,"tag":74,"props":568,"children":569},{},[570,575],{"type":48,"tag":78,"props":571,"children":572},{},[573],{"type":54,"value":574},"Report the result",{"type":54,"value":576}," — tell the user how many traces and spans were found and confirm tracing is working",{"type":48,"tag":578,"props":579,"children":581},"h3",{"id":580},"if-no-traces-appear",[582],{"type":54,"value":583},"If no traces appear",{"type":48,"tag":64,"props":585,"children":586},{},[587],{"type":54,"value":588},"Check these in order:",{"type":48,"tag":70,"props":590,"children":591},{},[592,618,644,670],{"type":48,"tag":74,"props":593,"children":594},{},[595,600,602,608,610,616],{"type":48,"tag":78,"props":596,"children":597},{},[598],{"type":54,"value":599},"Tracking URI not set",{"type":54,"value":601}," — is ",{"type":48,"tag":86,"props":603,"children":605},{"className":604},[],[606],{"type":54,"value":607},"mlflow.set_tracking_uri(...)",{"type":54,"value":609}," called before the agent run? Without this, traces go to a local ",{"type":48,"tag":86,"props":611,"children":613},{"className":612},[],[614],{"type":54,"value":615},".\u002Fmlruns",{"type":54,"value":617}," directory instead of the configured server.",{"type":48,"tag":74,"props":619,"children":620},{},[621,626,628,634,636,642],{"type":48,"tag":78,"props":622,"children":623},{},[624],{"type":54,"value":625},"Autolog warnings",{"type":54,"value":627}," — did ",{"type":48,"tag":86,"props":629,"children":631},{"className":630},[],[632],{"type":54,"value":633},"mlflow.autolog()",{"type":54,"value":635}," or framework-specific ",{"type":48,"tag":86,"props":637,"children":639},{"className":638},[],[640],{"type":54,"value":641},"mlflow.\u003Cframework>.autolog()",{"type":54,"value":643}," raise any warnings during setup? Check stderr for patching failures.",{"type":48,"tag":74,"props":645,"children":646},{},[647,652,654,660,662,668],{"type":48,"tag":78,"props":648,"children":649},{},[650],{"type":54,"value":651},"Wrong experiment ID",{"type":54,"value":653}," — verify the experiment ID passed to ",{"type":48,"tag":86,"props":655,"children":657},{"className":656},[],[658],{"type":54,"value":659},"search_traces()",{"type":54,"value":661}," matches the experiment active when the code ran (",{"type":48,"tag":86,"props":663,"children":665},{"className":664},[],[666],{"type":54,"value":667},"mlflow.get_experiment_by_name(...)",{"type":54,"value":669}," to confirm).",{"type":48,"tag":74,"props":671,"children":672},{},[673,678],{"type":48,"tag":78,"props":674,"children":675},{},[676],{"type":54,"value":677},"Network\u002Fauth issues",{"type":54,"value":679}," — can the process reach the tracking server? Check for connection errors or 401\u002F403 responses in logs.",{"type":48,"tag":64,"props":681,"children":682},{},[683,685,691],{"type":54,"value":684},"For automated validation, use ",{"type":48,"tag":86,"props":686,"children":688},{"className":687},[],[689],{"type":54,"value":690},"agent-evaluation\u002Fscripts\u002Fvalidate_tracing_runtime.py",{"type":54,"value":383},{"type":48,"tag":137,"props":693,"children":694},{},[],{"type":48,"tag":57,"props":696,"children":698},{"id":697},"feedback-collection",[699],{"type":54,"value":700},"Feedback Collection",{"type":48,"tag":64,"props":702,"children":703},{},[704],{"type":54,"value":705},"Log user feedback on traces for evaluation, debugging, and fine-tuning. Essential for identifying quality issues in production.",{"type":48,"tag":64,"props":707,"children":708},{},[709,711,717],{"type":54,"value":710},"See ",{"type":48,"tag":86,"props":712,"children":714},{"className":713},[],[715],{"type":54,"value":716},"references\u002Ffeedback-collection.md",{"type":54,"value":718}," for:",{"type":48,"tag":70,"props":720,"children":721},{},[722,733,738],{"type":48,"tag":74,"props":723,"children":724},{},[725,727],{"type":54,"value":726},"Recording user ratings and comments with ",{"type":48,"tag":86,"props":728,"children":730},{"className":729},[],[731],{"type":54,"value":732},"mlflow.log_feedback()",{"type":48,"tag":74,"props":734,"children":735},{},[736],{"type":54,"value":737},"Capturing trace IDs to return to clients",{"type":48,"tag":74,"props":739,"children":740},{},[741],{"type":54,"value":742},"LLM-as-judge automated evaluation",{"type":48,"tag":137,"props":744,"children":745},{},[],{"type":48,"tag":57,"props":747,"children":749},{"id":748},"reference-documentation",[750],{"type":54,"value":751},"Reference Documentation",{"type":48,"tag":578,"props":753,"children":755},{"id":754},"production-deployment",[756],{"type":54,"value":757},"Production Deployment",{"type":48,"tag":64,"props":759,"children":760},{},[761,762,768],{"type":54,"value":710},{"type":48,"tag":86,"props":763,"children":765},{"className":764},[],[766],{"type":54,"value":767},"references\u002Fproduction.md",{"type":54,"value":718},{"type":48,"tag":70,"props":770,"children":771},{},[772,777,782,787,800],{"type":48,"tag":74,"props":773,"children":774},{},[775],{"type":54,"value":776},"Environment variable configuration",{"type":48,"tag":74,"props":778,"children":779},{},[780],{"type":54,"value":781},"Async logging for low-latency applications",{"type":48,"tag":74,"props":783,"children":784},{},[785],{"type":54,"value":786},"Sampling configuration (MLFLOW_TRACE_SAMPLING_RATIO)",{"type":48,"tag":74,"props":788,"children":789},{},[790,792,798],{"type":54,"value":791},"Lightweight SDK (",{"type":48,"tag":86,"props":793,"children":795},{"className":794},[],[796],{"type":54,"value":797},"mlflow-tracing",{"type":54,"value":799},")",{"type":48,"tag":74,"props":801,"children":802},{},[803],{"type":54,"value":804},"Docker\u002FKubernetes deployment",{"type":48,"tag":578,"props":806,"children":808},{"id":807},"advanced-patterns",[809],{"type":54,"value":810},"Advanced Patterns",{"type":48,"tag":64,"props":812,"children":813},{},[814,815,821],{"type":54,"value":710},{"type":48,"tag":86,"props":816,"children":818},{"className":817},[],[819],{"type":54,"value":820},"references\u002Fadvanced-patterns.md",{"type":54,"value":718},{"type":48,"tag":70,"props":823,"children":824},{},[825,830,835],{"type":48,"tag":74,"props":826,"children":827},{},[828],{"type":54,"value":829},"Async function tracing",{"type":48,"tag":74,"props":831,"children":832},{},[833],{"type":54,"value":834},"Multi-threading with context propagation",{"type":48,"tag":74,"props":836,"children":837},{},[838],{"type":54,"value":839},"PII redaction with span processors",{"type":48,"tag":578,"props":841,"children":843},{"id":842},"distributed-tracing",[844],{"type":54,"value":845},"Distributed Tracing",{"type":48,"tag":64,"props":847,"children":848},{},[849,850,856],{"type":54,"value":710},{"type":48,"tag":86,"props":851,"children":853},{"className":852},[],[854],{"type":54,"value":855},"references\u002Fdistributed-tracing.md",{"type":54,"value":718},{"type":48,"tag":70,"props":858,"children":859},{},[860,865],{"type":48,"tag":74,"props":861,"children":862},{},[863],{"type":54,"value":864},"Propagating trace context across services",{"type":48,"tag":74,"props":866,"children":867},{},[868],{"type":54,"value":869},"Client\u002Fserver header APIs",{"type":48,"tag":578,"props":871,"children":873},{"id":872},"databricks-unity-catalog-storage",[874],{"type":54,"value":875},"Databricks (Unity Catalog storage)",{"type":48,"tag":64,"props":877,"children":878},{},[879,880,886,888,894],{"type":54,"value":710},{"type":48,"tag":86,"props":881,"children":883},{"className":882},[],[884],{"type":54,"value":885},"references\u002Fdatabricks.md",{"type":54,"value":887}," for storing traces in Unity Catalog Delta tables by binding an experiment to a ",{"type":48,"tag":86,"props":889,"children":891},{"className":890},[],[892],{"type":54,"value":893},"UnityCatalog",{"type":54,"value":895}," trace location (catalog, schema, table prefix).",{"type":48,"tag":897,"props":898,"children":899},"style",{},[900],{"type":54,"value":901},"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":903,"total":1056},[904,918,928,940,953,964,980,990,1003,1015,1030,1041],{"slug":905,"name":905,"fn":906,"description":907,"org":908,"tags":909,"stars":915,"repoUrl":916,"updatedAt":917},"setup","configure MLflow tracing","Configure MLflow tracing for Claude Code.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[910,913,914],{"name":911,"slug":912,"type":15},"Claude Code","claude-code",{"name":9,"slug":8,"type":15},{"name":13,"slug":14,"type":15},27014,"https:\u002F\u002Fgithub.com\u002Fmlflow\u002Fmlflow","2026-07-14T05:39:00.297769",{"slug":919,"name":919,"fn":920,"description":921,"org":922,"tags":923,"stars":915,"repoUrl":916,"updatedAt":927},"status","display MLflow tracing configuration","Show the current MLflow tracing configuration for Claude Code.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[924,925,926],{"name":911,"slug":912,"type":15},{"name":9,"slug":8,"type":15},{"name":13,"slug":14,"type":15},"2026-07-14T05:39:01.540537",{"slug":403,"name":403,"fn":929,"description":930,"org":931,"tags":932,"stars":32,"repoUrl":33,"updatedAt":939},"evaluate and optimize LLM agent output","Use this when you need to EVALUATE OR IMPROVE or OPTIMIZE an existing LLM agent's output quality - including improving tool selection accuracy, answer quality, reducing costs, or fixing issues where the agent gives wrong\u002Fincomplete responses. Evaluates agents systematically using MLflow evaluation with datasets, scorers, and tracing. IMPORTANT - Always also load the instrumenting-with-mlflow-tracing skill before starting any work. Covers end-to-end evaluation workflow or individual components (tracing setup, dataset creation, scorer definition, evaluation execution).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[933,934,935,936],{"name":30,"slug":31,"type":15},{"name":24,"slug":25,"type":15},{"name":9,"slug":8,"type":15},{"name":937,"slug":938,"type":15},"Performance","performance","2026-07-14T05:39:15.600492",{"slug":941,"name":941,"fn":942,"description":943,"org":944,"tags":945,"stars":32,"repoUrl":33,"updatedAt":952},"analyzing-mlflow-session","analyze MLflow chat conversation sessions","Analyzes an MLflow session — a sequence of traces from a multi-turn chat conversation or interaction. Use when the user asks to debug a chat conversation, review session or chat history, find where a multi-turn chat went wrong, or analyze patterns across turns. Triggers on \"analyze this session\", \"what happened in this conversation\", \"debug session\", \"review chat history\", \"where did this chat go wrong\", \"session traces\", \"analyze chat\", \"debug this chat\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[946,949,950,951],{"name":947,"slug":948,"type":15},"Debugging","debugging",{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},"2026-07-14T05:39:10.542342",{"slug":954,"name":954,"fn":955,"description":956,"org":957,"tags":958,"stars":32,"repoUrl":33,"updatedAt":963},"analyzing-mlflow-trace","analyze MLflow traces","Analyzes a single MLflow trace to answer a user query about it. Use when the user provides a trace ID and asks to debug, investigate, find issues, root-cause errors, understand behavior, or analyze quality. Triggers on \"analyze this trace\", \"what went wrong with this trace\", \"debug trace\", \"investigate trace\", \"why did this trace fail\", \"root cause this trace\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[959,960,961,962],{"name":947,"slug":948,"type":15},{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},"2026-07-14T05:39:02.874441",{"slug":965,"name":965,"fn":966,"description":967,"org":968,"tags":969,"stars":32,"repoUrl":33,"updatedAt":979},"fix-agent-issue","fix and update AI agent behavior","Drives a disciplined explore → plan → implement → verify loop for changing an AI agent's behavior with confidence — whether fixing a reported failure or introducing a new requirement, business rule, or policy. Grounds the diagnosis in MLflow traces, codifies the desired behavior as a regression test suite (`mlflow.genai.evaluate` assertions in `@mlflow.test` pytest tests), and iterates the agent — not the test — until green, resisting quick system-prompt patches when the real fix is upstream (missing tool, retrieval source, or capability). Use whenever the user wants to fix or change how an agent behaves — e.g. \"fix this issue in my agent\", \"this answer is wrong\", \"the agent is hallucinating\", \"improve my agent based on this trace\", \"make the agent do X instead of Y\", \"I want the agent to lead with\u002Fprioritize\u002Frecommend X\", \"new business rule: the agent should X\", \"always\u002Fnever do X\", \"change the agent's default behavior\" — or shares a trace they want addressed.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[970,971,974,975,978],{"name":30,"slug":31,"type":15},{"name":972,"slug":973,"type":15},"Best Practices","best-practices",{"name":947,"slug":948,"type":15},{"name":976,"slug":977,"type":15},"Engineering","engineering",{"name":9,"slug":8,"type":15},"2026-07-30T05:53:39.749464",{"slug":4,"name":4,"fn":5,"description":6,"org":981,"tags":982,"stars":32,"repoUrl":33,"updatedAt":34},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[983,984,985,986,987,988,989],{"name":30,"slug":31,"type":15},{"name":24,"slug":25,"type":15},{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":27,"slug":28,"type":15},{"name":13,"slug":14,"type":15},{"name":21,"slug":22,"type":15},{"slug":991,"name":991,"fn":992,"description":993,"org":994,"tags":995,"stars":32,"repoUrl":33,"updatedAt":1002},"mlflow-agent","dispatch MLflow workflows and agent tasks","Master dispatcher for all MLflow workflows. Use this skill when the user wants to do anything with MLflow — tracing, evaluating, debugging, or improving an agent. Routes to the right MLflow sub-skill automatically. Triggers on: \"use mlflow\", \"help with mlflow\", \"mlflow agent\", \"add mlflow to my project\", \"trace my agent\", \"evaluate my agent\", or any MLflow task without a specific skill in mind.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[996,997,998,1001],{"name":30,"slug":31,"type":15},{"name":9,"slug":8,"type":15},{"name":999,"slug":1000,"type":15},"MLOps","mlops",{"name":13,"slug":14,"type":15},"2026-07-14T05:39:04.133424",{"slug":1004,"name":1004,"fn":1005,"description":1006,"org":1007,"tags":1008,"stars":32,"repoUrl":33,"updatedAt":1014},"mlflow-onboarding","onboard users to MLflow workflows","Onboards users to MLflow by determining their use case (GenAI agents\u002Fapps or traditional ML\u002Fdeep learning) and guiding them through relevant quickstart tutorials and initial integration. If an experiment ID is available, it should be supplied as input to help determine the use case. Use when the user asks to get started with MLflow, set up tracking, add observability, or integrate MLflow into their project. Triggers on \"get started with MLflow\", \"set up MLflow\", \"onboard to MLflow\", \"add MLflow to my project\", \"how do I use MLflow\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1009,1010,1011],{"name":9,"slug":8,"type":15},{"name":999,"slug":1000,"type":15},{"name":1012,"slug":1013,"type":15},"Onboarding","onboarding","2026-07-14T05:39:06.681003",{"slug":1016,"name":1016,"fn":1017,"description":1018,"org":1019,"tags":1020,"stars":32,"repoUrl":33,"updatedAt":1029},"querying-mlflow-metrics","fetch trace metrics from MLflow servers","Fetches aggregated trace metrics (token usage, latency, trace counts, quality evaluations) from MLflow tracking servers. Triggers on requests to show metrics, analyze token usage, view LLM costs, check usage trends, or query trace statistics.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1021,1024,1027,1028],{"name":1022,"slug":1023,"type":15},"Analytics","analytics",{"name":1025,"slug":1026,"type":15},"Metrics","metrics",{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},"2026-07-14T05:39:13.07056",{"slug":1031,"name":1031,"fn":1032,"description":1033,"org":1034,"tags":1035,"stars":32,"repoUrl":33,"updatedAt":1040},"retrieving-mlflow-traces","retrieve and query MLflow traces","Retrieves MLflow traces using CLI or Python API. Use when the user asks to get a trace by ID, find traces, filter traces by status\u002Ftags\u002Fmetadata\u002Fexecution time, query traces, or debug failed traces. Triggers on \"get trace\", \"search traces\", \"find failed traces\", \"filter traces by\", \"traces slower than\", \"query MLflow traces\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1036,1037,1038,1039],{"name":947,"slug":948,"type":15},{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},"2026-07-14T05:39:09.22888",{"slug":1042,"name":1042,"fn":1043,"description":1044,"org":1045,"tags":1046,"stars":32,"repoUrl":33,"updatedAt":1055},"sagemaker-mlflow","connect to SageMaker Managed MLflow","Connect to SageMaker Managed MLflow (mlflow-app or mlflow-tracking-server ARN) as an MLflow backend, then hand off to the other MLflow skills. Triggers on a SageMaker MLflow ARN (arn:aws:sagemaker:...:mlflow-app\u002F... or arn:aws:sagemaker:...:mlflow-tracking-server\u002F...) or \"SageMaker Managed MLflow\".\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1047,1050,1053,1054],{"name":1048,"slug":1049,"type":15},"AI Infrastructure","ai-infrastructure",{"name":1051,"slug":1052,"type":15},"AWS","aws",{"name":9,"slug":8,"type":15},{"name":999,"slug":1000,"type":15},"2026-07-14T05:39:05.401801",13,{"items":1058,"total":1111},[1059,1066,1073,1080,1088,1098,1105],{"slug":403,"name":403,"fn":929,"description":930,"org":1060,"tags":1061,"stars":32,"repoUrl":33,"updatedAt":939},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1062,1063,1064,1065],{"name":30,"slug":31,"type":15},{"name":24,"slug":25,"type":15},{"name":9,"slug":8,"type":15},{"name":937,"slug":938,"type":15},{"slug":941,"name":941,"fn":942,"description":943,"org":1067,"tags":1068,"stars":32,"repoUrl":33,"updatedAt":952},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1069,1070,1071,1072],{"name":947,"slug":948,"type":15},{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},{"slug":954,"name":954,"fn":955,"description":956,"org":1074,"tags":1075,"stars":32,"repoUrl":33,"updatedAt":963},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1076,1077,1078,1079],{"name":947,"slug":948,"type":15},{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},{"slug":965,"name":965,"fn":966,"description":967,"org":1081,"tags":1082,"stars":32,"repoUrl":33,"updatedAt":979},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1083,1084,1085,1086,1087],{"name":30,"slug":31,"type":15},{"name":972,"slug":973,"type":15},{"name":947,"slug":948,"type":15},{"name":976,"slug":977,"type":15},{"name":9,"slug":8,"type":15},{"slug":4,"name":4,"fn":5,"description":6,"org":1089,"tags":1090,"stars":32,"repoUrl":33,"updatedAt":34},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1091,1092,1093,1094,1095,1096,1097],{"name":30,"slug":31,"type":15},{"name":24,"slug":25,"type":15},{"name":9,"slug":8,"type":15},{"name":17,"slug":18,"type":15},{"name":27,"slug":28,"type":15},{"name":13,"slug":14,"type":15},{"name":21,"slug":22,"type":15},{"slug":991,"name":991,"fn":992,"description":993,"org":1099,"tags":1100,"stars":32,"repoUrl":33,"updatedAt":1002},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1101,1102,1103,1104],{"name":30,"slug":31,"type":15},{"name":9,"slug":8,"type":15},{"name":999,"slug":1000,"type":15},{"name":13,"slug":14,"type":15},{"slug":1004,"name":1004,"fn":1005,"description":1006,"org":1106,"tags":1107,"stars":32,"repoUrl":33,"updatedAt":1014},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1108,1109,1110],{"name":9,"slug":8,"type":15},{"name":999,"slug":1000,"type":15},{"name":1012,"slug":1013,"type":15},11]