docs: document provider SDK auto-instrumentation

docs.json

@@ -89,28 +89,53 @@
             "group": "Observability",
             "icon": "chart-column",
             "pages": [
-              "features/observability",
-              "why-promptlayer/advanced-search",
-              "why-promptlayer/analytics",
-              "running-requests/traces",
-              "features/opentelemetry",
-              "features/integrations",
+              "features/observability/overview",
               {
-                "group": "Advanced Logging",
-                "icon": "cassette-tape",
+                "group": "Request Logs",
+                "icon": "rectangle-list",
                 "pages": [
-                  "features/prompt-history/request-id",
-                  "features/prompt-history/tagging-requests",
-                  "features/prompt-history/metadata",
-                  "features/prompt-history/scoring-requests",
-                  "features/prompt-history/tracking-templates",
-                  "features/prompt-history/custom-logging",
-                  "features/prompt-history/structured-output-logging",
-                  "features/prompt-history/search-data-model"
+                  "features/observability/request-logs",
+                  "features/observability/request-logs/request-ids",
+                  "features/observability/request-logs/tags",
+                  "features/observability/request-logs/metadata",
+                  "features/observability/request-logs/scores",
+                  "features/observability/request-logs/tracking-templates",
+                  "features/observability/request-logs/custom-logging",
+                  "features/observability/request-logs/structured-outputs"
+                ]
+              },
+              {
+                "group": "Traces",
+                "icon": "diagram-project",
+                "pages": [
+                  "features/observability/traces",
+                  "features/observability/traces/manual-tracing",
+                  {
+                    "group": "SDK Auto-Instrumentation",
+                    "icon": "code",
+                    "pages": [
+                      "features/observability/traces/auto-instrumentation/overview",
+                      "features/observability/traces/auto-instrumentation/openai",
+                      "features/observability/traces/auto-instrumentation/anthropic",
+                      "features/observability/traces/auto-instrumentation/google",
+                      "features/observability/traces/auto-instrumentation/bedrock"
+                    ]
+                  },
+                  "features/observability/traces/opentelemetry",
+                  "features/observability/traces/integrations"
                 ]
               }
             ]
           },
+          {
+            "group": "Search & Analytics",
+            "icon": "chart-pie-simple",
+            "pages": [
+              "features/search-and-analytics/search",
+              "features/search-and-analytics/analytics",
+              "features/search-and-analytics/search-data-model"
+            ]
+          },
           "why-promptlayer/workflows",
           {
             "group": "Tool Registry",
@@ -512,6 +537,62 @@
     }
   },
   "redirects": [
+    {
+      "source": "/features/observability",
+      "destination": "/features/observability/overview"
+    },
+    {
+      "source": "/running-requests/traces",
+      "destination": "/features/observability/traces"
+    },
+    {
+      "source": "/features/integrations",
+      "destination": "/features/observability/traces/integrations"
+    },
+    {
+      "source": "/features/opentelemetry",
+      "destination": "/features/observability/traces/opentelemetry"
+    },
+    {
+      "source": "/features/prompt-history/request-id",
+      "destination": "/features/observability/request-logs/request-ids"
+    },
+    {
+      "source": "/features/prompt-history/tagging-requests",
+      "destination": "/features/observability/request-logs/tags"
+    },
+    {
+      "source": "/features/prompt-history/metadata",
+      "destination": "/features/observability/request-logs/metadata"
+    },
+    {
+      "source": "/features/prompt-history/scoring-requests",
+      "destination": "/features/observability/request-logs/scores"
+    },
+    {
+      "source": "/features/prompt-history/tracking-templates",
+      "destination": "/features/observability/request-logs/tracking-templates"
+    },
+    {
+      "source": "/features/prompt-history/custom-logging",
+      "destination": "/features/observability/request-logs/custom-logging"
+    },
+    {
+      "source": "/features/prompt-history/structured-output-logging",
+      "destination": "/features/observability/request-logs/structured-outputs"
+    },
+    {
+      "source": "/features/prompt-history/search-data-model",
+      "destination": "/features/search-and-analytics/search-data-model"
+    },
+    {
+      "source": "/why-promptlayer/advanced-search",
+      "destination": "/features/search-and-analytics/search"
+    },
+    {
+      "source": "/why-promptlayer/analytics",
+      "destination": "/features/search-and-analytics/analytics"
+    },
     {
       "source": "/sdks/evals/setup-with-ai",
       "destination": "/sdks/evals/quickstart"

features/faq.mdx

@@ -15,7 +15,7 @@ Yes, PromptLayer supports multi-modal image models, including `gpt-4-vision-prev
 To use `gpt-4-vision-preview` with PromptLayer, follow these steps:
 
 1. Ensure you have the PromptLayer and OpenAI Python libraries installed.
-2. Use the [`run()` method](/sdks/python#using-the-run-method-recommended) to execute prompts, or use [`log_request`](/features/prompt-history/custom-logging) to log requests made with your own client.
+2. Use the [`run()` method](/sdks/python#using-the-run-method-recommended) to execute prompts, or use [`log_request`](/features/observability/request-logs/custom-logging) to log requests made with your own client.
 3. Make your request to `gpt-4-vision-preview` with the necessary image inputs, either through image URLs or base64 encoded images.
 4. Check the PromptLayer dashboard to see your request logged!
 
@@ -25,7 +25,7 @@ Multi-modal models are also supported in the Prompt Registry, Playground, and Ev
 
 ## Do you support OpenAI function calling?
 
-Yes, we take great pride in staying up to date. PromptLayer supports [function calling](https://platform.openai.com/docs/guides/function-calling) through the `run()` method and via [custom logging](/features/prompt-history/custom-logging). You can also configure tool calling directly in the [Prompt Registry](/features/prompt-registry/tool-calling).
+Yes, we take great pride in staying up to date. PromptLayer supports [function calling](https://platform.openai.com/docs/guides/function-calling) through the `run()` method and via [custom logging](/features/observability/request-logs/custom-logging). You can also configure tool calling directly in the [Prompt Registry](/features/prompt-registry/tool-calling).
 
 
 ## Does PromptLayer support streaming?
@@ -135,7 +135,7 @@ PromptLayer provides out-of-the-box support for Mistral in our logs, playground,
 
 ## What's the difference between tags and metadata?
 
-Both [tags](/features/prompt-history/tagging-requests) and [metadata](/features/prompt-history/metadata) enable the addition of supplementary information to your request logs, yet they serve distinct purposes. Tags are ideal for classifying requests into a limited number of predefined categories, such as "prod" or "dev". Conversely, metadata is tailored for capturing unique, request-specific details like user IDs or session IDs.
+Both [tags](/features/observability/request-logs/tags) and [metadata](/features/observability/request-logs/metadata) enable the addition of supplementary information to your request logs, yet they serve distinct purposes. Tags are ideal for classifying requests into a limited number of predefined categories, such as "prod" or "dev". Conversely, metadata is tailored for capturing unique, request-specific details like user IDs or session IDs.
 
 ## Why do I see extra input variables in my prompt template? Parsing does not seem to be working.
 

features/image-generation.mdx

@@ -337,7 +337,7 @@ PromptLayer automatically handles image storage for generated images:
 
 ## Logging Image Generation Requests
 
-If you're making image generation calls with your own client, you can log them to PromptLayer using [`log_request`](/features/prompt-history/custom-logging). PromptLayer recognizes the following function names for image generation:
+If you're making image generation calls with your own client, you can log them to PromptLayer using [`log_request`](/features/observability/request-logs/custom-logging). PromptLayer recognizes the following function names for image generation:
 
 - `openai.images.generate`
 - `openai.OpenAI.images.generate`
@@ -354,4 +354,4 @@ Image generation outputs work with PromptLayer's evaluation system. Generated im
 - [Tool Calling (Built-in Tools)](/features/prompt-registry/tool-calling)
 - [Python SDK Run Method](/sdks/python#using-the-run-method-recommended)
 - [JavaScript SDK Run Method](/sdks/javascript#using-the-run-method-recommended)
-- [Custom Logging](/features/prompt-history/custom-logging)
+- [Custom Logging](/features/observability/request-logs/custom-logging)

features/observability/overview.mdx

@@ -0,0 +1,72 @@
+---
+title: "Overview"
+description: "Understand the request logs and traces that make up PromptLayer observability."
+icon: "book"
+---
+
+Observability helps you inspect model calls and follow the execution of AI applications across prompts, agents, tools, and workflows.
+
+PromptLayer organizes observability data into two artifacts:
+
+| Artifact | Use it to |
+| --- | --- |
+| [**Request log**](/features/observability/request-logs) | Inspect one model call, including its input, output, model, timing, tokens, cost, status, metadata, and prompt association. |
+| [**Trace**](/features/observability/traces) | Follow one end-to-end operation as a hierarchy of LLM calls, agents, tools, retrieval steps, and custom application spans. |
+
+A trace can contain multiple model calls and therefore link to multiple request logs. Supported direct GenAI spans create request logs, while spans managed by `PromptLayer.run()` can link to the run's existing request log.
+
+## Where to Go Next
+
+<CardGroup cols={2}>
+  <Card
+    title="Request Logs"
+    icon="rectangle-list"
+    href="/features/observability/request-logs"
+  >
+    Review individual model calls, errors, inputs, outputs, usage, and cost.
+  </Card>
+  <Card
+    title="Traces"
+    icon="diagram-project"
+    href="/features/observability/traces"
+  >
+    Inspect the parent-child execution path across model calls, tools, agents, and application code.
+  </Card>
+  <Card
+    title="Search"
+    icon="magnifying-glass"
+    href="/features/search-and-analytics/search"
+  >
+    Find request logs or traces with content queries, structured filters, and date ranges.
+  </Card>
+  <Card
+    title="Analytics"
+    icon="chart-pie-simple"
+    href="/features/search-and-analytics/analytics"
+  >
+    Aggregate the current request-log or trace query into charts.
+  </Card>
+</CardGroup>
+
+## Configure Observability
+
+Once you know which artifact you need, follow the setup path for the code that produces it:
+
+<CardGroup cols={2}>
+  <Card
+    title="Configure Request Logging"
+    icon="rectangle-list"
+    href="/features/observability/request-logs"
+  >
+    Choose how requests are captured, then add IDs, tags, metadata, scores, and prompt associations.
+  </Card>
+  <Card
+    title="Configure Tracing"
+    icon="diagram-project"
+    href="/features/observability/traces"
+  >
+    Choose PromptLayer SDK spans, provider auto-instrumentation, a telemetry integration, or OpenTelemetry.
+  </Card>
+</CardGroup>
+
+For indexing details, see the [Search Data Model](/features/search-and-analytics/search-data-model).

features/observability/request-logs.mdx

@@ -0,0 +1,57 @@
+---
+title: "Overview"
+description: "Inspect individual model calls, including their content, status, usage, cost, and metadata."
+icon: "rectangle-list"
+---
+
+A request log records one model call. Use it to inspect exactly what your application sent and received, diagnose a failure, review latency and token usage, or connect production behavior to a prompt template.
+
+Each request log can include:
+
+- Model input and output
+- Provider and model
+- Start time, end time, and latency
+- Input and output tokens
+- Estimated cost
+- Status and error details
+- Prompt template association
+- Tags, metadata, scores, and a PromptLayer request ID
+
+## View Request Logs
+
+Open **Request Logs** in the PromptLayer sidebar. The Requests table shows the model, prompt, status, timing, token usage, cost, and other fields for each call. Select a row to open its full details.
+
+<Frame>
+  <img src="/new-quickstart-images/logs-table.png" alt="PromptLayer request logs table" />
+</Frame>
+
+To review logs for one prompt, open that prompt and select **Analytics & Logs**.
+
+<Frame>
+  <img
+    src="/new-quickstart-images/logs-by-prompt-display.png"
+    alt="Request logs filtered to one prompt"
+  />
+</Frame>
+
+## Find a Request
+
+Use the date range, free-text query, and structured filters above the Requests table. You can filter by fields such as prompt, model, provider, status, tags, metadata, content, latency, tokens, and cost.
+
+See [Search](/features/search-and-analytics/search) for the dashboard workflow, supported filter categories, and older-data behavior.
+
+## Request Logs and Traces
+
+A request log describes one model call. A [trace](/features/observability/traces) describes the complete operation around that call and can contain multiple request logs.
+
+When PromptLayer receives a supported GenAI span, it creates a request log for the model call and links it to the trace. If `PromptLayer.run()` already created the request log, PromptLayer links the span to that log instead of creating a duplicate.
+
+## Configure Request Logs
+
+Use [tags](/features/observability/request-logs/tags), [metadata](/features/observability/request-logs/metadata), and [scores](/features/observability/request-logs/scores) to make requests easier to find and compare. Every log also has a [PromptLayer request ID](/features/observability/request-logs/request-ids) that you can store or use with tracking methods.
+
+Use [Tracking Templates](/features/observability/request-logs/tracking-templates) to associate provider calls with PromptLayer prompts. For manual provider calls or unsupported models, use [Custom Logging](/features/observability/request-logs/custom-logging). If a provider returns JSON, see [Structured Outputs](/features/observability/request-logs/structured-outputs).
+
+## Reuse Production Data
+
+Select request logs from the table and add them to a [PromptLayer Table](/features/tables/overview#import-data). This lets you turn representative production examples or failures into evaluation and backtesting data.

features/prompt-history/custom-logging.mdx → features/observability/request-logs/custom-logging.mdx

@@ -4,6 +4,8 @@ icon: "brackets-curly"
 ---
 
 
+Custom logging creates [request logs](/features/observability/request-logs) for calls that PromptLayer does not capture automatically. Return to [Configure Request Logging](/features/observability/request-logs) to compare capture and enrichment options.
+
 ## When to Use Custom Logging
 
 Use the `log_request` method when: