WEAVE: Integrates feedback for Weave Agents GA (#2778)

## Description
This updates the Getting Started section and landing page to be more
Agent-centric and produce a smoother onboarding experience.

---------

Co-authored-by: anastasiaguspan <aguspan@coreweave.com>
Co-authored-by: wandb-docs-pr-writer[bot] <281709622+wandb-docs-pr-writer[bot]@users.noreply.github.com>

Author: 3 people

docs.json

@@ -531,8 +531,8 @@
                   {
                     "group": "Get Started",
                     "pages": [
-                      "weave/agents-quickstart",
-                      "weave/quickstart"
+                      "weave/agent-integration-quickstart",
+                      "weave/custom-agents-quickstart"
                     ]
                   },
                   {
@@ -546,6 +546,7 @@
                           "weave/guides/tracking/trace-agent-integrations",
                           "weave/guides/tracking/view-agent-activity",
                           "weave/guides/tracking/view-agent-signals",
+                          "weave/guides/tracking/trace-agents-attributes",
                           "weave/guides/tracking/trace-agents-batch",
                           "weave/guides/tracking/trace-sub-agents"
                         ]
@@ -556,6 +557,7 @@
                           {
                             "group": "Tracing basics",
                             "pages": [
+                              "weave/quickstart",
                               "weave/guides/tracking/tracing",
                               "weave/guides/tracking/create-call",
                               "weave/guides/tracking/trace-tree",
@@ -687,18 +689,14 @@
                           {
                             "group": "Frameworks",
                             "pages": [
-                              "weave/guides/integrations/openai_agents",
                               "weave/guides/integrations/openai-realtime-audio",
-                              "weave/guides/integrations/claude_agent",
-                              "weave/guides/integrations/claude_code",
                               "weave/guides/integrations/langchain",
                               "weave/guides/integrations/llamaindex",
                               "weave/guides/integrations/dspy",
                               "weave/guides/integrations/instructor",
                               "weave/guides/integrations/crewai",
                               "weave/guides/integrations/smolagents",
                               "weave/guides/integrations/pydantic_ai",
-                              "weave/guides/integrations/google_adk",
                               "weave/guides/integrations/vercel_ai_sdk",
                               "weave/guides/integrations/haystack",
                               "weave/guides/integrations/agno",
@@ -1997,7 +1995,6 @@
                   {
                     "group": "Premiers pas",
                     "pages": [
-                      "fr/weave/agents-quickstart",
                       "fr/weave/quickstart"
                     ]
                   },
@@ -3382,7 +3379,6 @@
                   {
                     "group": "スタートガイド",
                     "pages": [
-                      "ja/weave/agents-quickstart",
                       "ja/weave/quickstart"
                     ]
                   },
@@ -4767,7 +4763,6 @@
                   {
                     "group": "시작하기",
                     "pages": [
-                      "ko/weave/agents-quickstart",
                       "ko/weave/quickstart"
                     ]
                   },
@@ -7060,6 +7055,22 @@
       "destination": "/weave/guides/evaluation/monitors",
       "source": "/weave/guides/evaluation/guardrails_and_monitors"
     },
+    {
+      "destination": "/weave/guides/integrations/agents/claude-agents-sdk",
+      "source": "/weave/guides/integrations/claude_agent"
+    },
+    {
+      "destination": "/weave/guides/integrations/agents/claude-code-harness",
+      "source": "/weave/guides/integrations/claude_code"
+    },
+    {
+      "destination": "/weave/guides/integrations/agents/google-adk",
+      "source": "/weave/guides/integrations/google_adk"
+    },
+    {
+      "destination": "/weave/guides/integrations/agents/openai-agents-sdk",
+      "source": "/weave/guides/integrations/openai_agents"
+    },
     {
       "destination": "weave/concepts/what-is-weave",
       "source": "weave/concepts/fundamental-components"

snippets/_includes/agent-integration-paths.mdx

@@ -0,0 +1,23 @@
+## Agent harnesses
+
+Agent harnesses are end-user agent runtimes, such as coding agents and developer tools. Install the plugin or extension, then start a session in the harness to begin tracing:
+
+- [Claude Code plugin](/weave/guides/integrations/agents/claude-code-harness)
+- [Codex plugin](/weave/guides/integrations/agents/codex-harness)
+- [OpenClaw plugin](/weave/guides/integrations/agents/openclaw-harness)
+- [Pi extension](/weave/guides/integrations/agents/pi-dev-harness)
+
+## Agent SDKs
+
+Agent SDKs are libraries for building agents and multi-agent workflows in your own application code. Add `weave.init("[YOUR-TEAM]/[YOUR-PROJECT]")` to your script and Weave autopatches the SDK:
+
+- [OpenAI Agents SDK](/weave/guides/integrations/agents/openai-agents-sdk)
+- [Google Agent Development Kit (ADK)](/weave/guides/integrations/agents/google-adk)
+- [Claude Agent SDK](/weave/guides/integrations/agents/claude-agents-sdk)
+
+## Custom agents and OpenTelemetry
+
+If your agent isn't built with one of the SDKs above, or it already emits OpenTelemetry spans, use the Weave SDK directly. Weave accepts any OTel span and gives special handling to [GenAI semantic-convention attributes](https://opentelemetry.io/docs/specs/semconv/gen-ai/gen-ai-agent-spans/) so spans render in the **Agents** view:
+
+- [Quickstart: Manually instrument an agent](/weave/custom-agents-quickstart) — a walkthrough of a small multi-turn agent.
+- [Trace your agents](/weave/guides/tracking/trace-agents) — the full reference of agent helpers and concepts.

weave.mdx

@@ -4,9 +4,10 @@ description: "Track, test, and improve language model apps with W&B Weave"
 mode: wide
 ---
 
-W&B Weave is an observability and evaluation platform that helps you track, evaluate, and improve your LLM application. With Weave, you can:
+W&B Weave is an observability and evaluation platform that helps you track, evaluate, and improve your agents and LLM applications. With Weave, you can:
 
-* [Observe and debug](/weave/quickstart) your LLM application
+* [Trace and collect metrics about your agent built with popular SDKs and harnesses](/weave/agent-integration-quickstart)
+* [Manually instrument custom-built agents to collect traces and metrics](/weave/custom-agents-quickstart)
 * [Evaluate](/weave/tutorial-eval) your application’s responses using LLM judges and custom scorers
 
 ## Get started
@@ -15,28 +16,29 @@ The following docs guide you through the basics of how to use Weave's suite of t
 
 <CardGroup cols={3}>
   <Card
-    title="Quickstart: Track LLM inputs & outputs"
-    icon="chart-line"
-    href="/weave/quickstart"
+    title="Integrate Weave with an agent"
+    icon="plug"
+    href="/weave/agent-integration-quickstart"
   >
-    Start by tracing a basic call to an LLM and reviewing the data in your W&B account.
+    Pick a built-in integration for your agent SDK or harness and start tracing sessions, turns, LLM calls, and tool calls in the Agents view.
   </Card>
 
   <Card
-    title="Get started evaluating your app"
-    icon="clipboard-check"
-    href="/weave/tutorial-eval"
+    title="Set up custom agent observability"
+    icon="code"
+    href="/weave/custom-agents-quickstart"
   >
-    Learn how to build an evaluation pipeline using Weave scorers to test and track your application's performance.
+    Trace a custom multi-turn agent with the Weave SDK to capture sessions, LLM calls, and tool calls as OpenTelemetry spans.
   </Card>
 
   <Card
-    title="Evaluate a RAG application"
-    icon="search"
-    href="/weave/tutorial-rag"
+    title="View agent metrics in Weave UI"
+    icon="chart-line"
+    href="/weave/guides/tracking/view-agent-activity"
   >
-    Build and evaluate RAG applications using Weave with LLM judges to measure retrieval quality.
+    Trace a basic call to an LLM and review the inputs, outputs, and code in your W&B account.
   </Card>
+
 </CardGroup>
 
 ## Install Weave

weave/agent-integration-quickstart.mdx

@@ -0,0 +1,20 @@
+---
+title: "Choose an agent integration"
+description: Pick a built-in Weave integration for your agent SDK or harness and start sending sessions, turns, LLM calls, and tool calls to the Agents view.
+keywords: ["agent integrations", "autopatch", "agent SDKs", "agent harnesses", "tracing"]
+---
+
+import AgentsPreview from '/snippets/_includes/agents-public-preview.mdx';
+import AgentIntegrationPaths from '/snippets/_includes/agent-integration-paths.mdx';
+
+<AgentsPreview />
+
+W&B Weave traces multi-turn agents built with popular SDKs and harnesses without hand-instrumenting each turn. Install a plugin for your agent harness, or call `weave.init()` in code that uses a supported agent SDK, and Weave autopatches the framework. Sessions, turns, LLM calls, and tool calls render in the **Agents** view of your project.
+
+Choose from one of the following supported integrations.
+
+<AgentIntegrationPaths />
+
+## Next steps
+
+Once your first session reaches Weave, see [View activity in the Agents view](/weave/guides/tracking/view-agent-activity) to explore sessions, turns, and tool calls.

weave/concepts/what-is-weave.mdx

@@ -23,7 +23,7 @@ Weave provides the following core functionality:
 
 Weave provides agentic observability for the full lifecycle of agent conversations, including sessions, LLM calls, and tool executions.
 
-If you're building an agent, follow the [agent tracing quickstart](weave/agents-quickstart.mdx) or learn to use the Weave SDK to [trace your agents](/weave/guides/tracking/trace-agents).
+If you're building an agent, start with the [agent integration quickstart](/weave/agent-integration-quickstart) to pick a built-in integration, or follow the [custom agents quickstart](/weave/custom-agents-quickstart) to manually instrument your own. To learn how the Weave SDK models agents, see [Trace your agents](/weave/guides/tracking/trace-agents).
 
 If you're using a supported third-party agent harness, such as Claude Code or OpenAI Agent SDK, Weave instruments it automatically with no additional code. See [Integrations](/weave/guides/integrations) for all supported frameworks.
 

weave/agents-quickstart.mdx weave/custom-agents-quickstart.mdxweave/agents-quickstart.mdx renamed to weave/custom-agents-quickstart.mdx

@@ -1,5 +1,5 @@
 ---
-title: "Quickstart: Trace an agent"
+title: "Quickstart: Set up custom agent observability"
 description: Trace a multi-turn agent with the Weave SDK. Sessions, turns, LLM calls, and tool calls render in the Agents view of your project.
 keywords: ["OpenTelemetry", "OTel spans", "multi-turn agent", "tool calls", "tracing"]
 ---

weave/guides/evaluation/custom-monitors.mdx

@@ -3,6 +3,10 @@ title: "Set up custom monitors"
 description: "Passively score production traffic to surface trends and issues"
 ---
 
+<Note>
+Custom monitors are the previous approach to monitoring production traffic. For new implementations, use **Signals** under Weave for Agents. See [View agent signals](/weave/guides/tracking/view-agent-signals).
+</Note>
+
 This page shows you how to set up custom monitors in W&B Weave so you can passively score production traffic and surface trends and issues in your LLM applications. Use this guide when you want to define your own evaluation criteria, beyond Weave's preset signals, for production traces.
 
 Monitors use LLM judges to passively score production traffic to surface trends and issues in your LLM applications. For example, you can monitor your application's responses for correctness or helpfulness, or you can monitor user input to identify trends in what users ask your agents about. Monitors automatically store all scoring results in Weave's database, so you can analyze historical trends and patterns.

weave/guides/evaluation/monitors.mdx

@@ -5,7 +5,7 @@ keywords: ["signals", "monitors", "LLM-as-a-judge", "hallucination", "production
 ---
 
 <Note>
-When tracing agentic applications, use the **Signals** tab on the **Agents** page. See [Signals for agents](/weave/guides/tracking/view-agent-signals).
+This page covers a previous approach to monitoring production traffic. For new implementations, use **Signals** under Weave for Agents. See [View agent signals](/weave/guides/tracking/view-agent-signals).
 </Note>
 
 

weave/guides/integrations.mdx

@@ -46,9 +46,9 @@ LLM providers are the vendors that offer access to large language models to gene
 
 Frameworks help orchestrate the execution pipelines in AI applications. They provide tools and abstractions for building complex workflows. Weave integrates with these frameworks to trace the entire pipeline:
 
-- **[OpenAI Agents SDK (call-level tracing)](/weave/guides/integrations/openai_agents)**
-- **[Claude Agent SDK (call-level tracing)](/weave/guides/integrations/claude_agent)**
-- **[Claude Code (call-level tracing)](/weave/guides/integrations/claude_code)**
+- **[OpenAI Agents SDK](/weave/guides/integrations/agents/openai-agents-sdk)**
+- **[Claude Agent SDK](/weave/guides/integrations/agents/claude-agents-sdk)**
+- **[Claude Code](/weave/guides/integrations/agents/claude-code-harness)**
 - **[LangChain](/weave/guides/integrations/langchain)**
 - **[LlamaIndex](/weave/guides/integrations/llamaindex)**
 - **[Haystack](/weave/guides/integrations/haystack)**
@@ -57,7 +57,7 @@ Frameworks help orchestrate the execution pipelines in AI applications. They pro
 - **[CrewAI](/weave/guides/integrations/crewai)**
 - **[Smolagents](/weave/guides/integrations/smolagents)**
 - **[PydanticAI](/weave/guides/integrations/pydantic_ai)**
-- **[Google Agent Development Kit (ADK, call-level tracing)](/weave/guides/integrations/google_adk)**
+- **[Google Agent Development Kit (ADK)](/weave/guides/integrations/agents/google-adk)**
 - **[AutoGen](/weave/guides/integrations/autogen)**
 - **[Verdict](/weave/guides/integrations/verdict)**
 - **[TypeScript SDK](/weave/guides/integrations/js)**