diff --git a/docs/ai/index.mdx b/docs/ai/index.mdx index f3cfec913828b..b73e5af58ec0d 100644 --- a/docs/ai/index.mdx +++ b/docs/ai/index.mdx @@ -45,9 +45,9 @@ keywords: className="w-full md:w-[calc(50%-12px)]" /> diff --git a/docs/ai/observability/agents/index.mdx b/docs/ai/observability/agents/index.mdx deleted file mode 100644 index 7ca427be75a4f..0000000000000 --- a/docs/ai/observability/agents/index.mdx +++ /dev/null @@ -1,34 +0,0 @@ ---- -title: AI Agents -sidebar_order: 10 -description: >- - Learn how to use Sentry's Agent tracing tools to trace and debug your AI - agent workflows, including agent runs, tool calls, and model interactions. -keywords: - - AI agents - - LLM monitoring - - agent tracing - - tool calls - - OpenAI - - Anthropic - - LangChain - - token usage - - agent tracing -og_image: /og-images/ai-monitoring-agents.png ---- - -Sentry's Agent tracing tools help you understand what's going on with your AI agent workflows. They automatically collect information about agent runs, tool calls, model interactions, and errors across your entire AI pipeline—from user interaction to final response. - -## Example Agent Tracing Use Cases - -- Your AI agent is failing silently during tool execution, and you want to trace the complete agent flow to identify where it's breaking. -- Users report that your AI agent is returning unexpected or malformed responses, and you need to debug the full context of prompts, model calls, and outputs. -- Your AI agent workflows are experiencing performance issues, and you want to identify which steps (model calls, tool usage, or custom logic) are causing bottlenecks. - -## Get Started - -To use Agent Tracing, you must have an existing Sentry account and project set up. If you don't have one, [create an account here](https://sentry.io/signup/). - -![Agents Tracing Overview](./img/overview.png) - -Learn how to [set up Sentry for AI Agents](/ai/observability/agents/getting-started/) and [name your agents](/ai/observability/agents/naming/) so they're identifiable in the dashboard. diff --git a/docs/ai/observability/index.mdx b/docs/ai/observability/index.mdx deleted file mode 100644 index 679d25bf7cc0b..0000000000000 --- a/docs/ai/observability/index.mdx +++ /dev/null @@ -1,39 +0,0 @@ ---- -title: AI Observability -sidebar_order: 30 -description: "Monitor your AI agents and MCP servers in production with Sentry. Track LLM calls, token usage, tool executions, and agent handoffs." -keywords: - - AI observability - - LLM monitoring - - agent tracing - - MCP monitoring - - token usage - - AI observability - - AI tracing ---- - -Sentry provides comprehensive monitoring for your AI implementations. Whether you're building AI agents or MCP servers, Sentry automatically captures performance metrics, errors, and traces across your entire AI pipeline. - - - - - - diff --git a/docs/guides/custom-spans.mdx b/docs/guides/custom-spans.mdx index c1a2952c8257a..d90da0f6f3eb1 100644 --- a/docs/guides/custom-spans.mdx +++ b/docs/guides/custom-spans.mdx @@ -638,7 +638,7 @@ fun processEmailDigest(job: Job) { ### 5. AI/LLM Operations -For AI workloads, use [Sentry Agent Tracing](/ai/observability/agents/) instead of manual instrumentation when possible. It automatically captures agent workflows, tool calls, and token usage. +For AI workloads, use [Sentry Agent Tracing](/product/agents/) instead of manual instrumentation when possible. It automatically captures agent workflows, tool calls, and token usage. If you're not using a supported framework or need custom attributes: diff --git a/docs/platforms/dotnet/common/tracing/instrumentation/ai-agents-module.mdx b/docs/platforms/dotnet/common/tracing/instrumentation/ai-agents-module.mdx index 61c9102199fcf..f27f76c4ae107 100644 --- a/docs/platforms/dotnet/common/tracing/instrumentation/ai-agents-module.mdx +++ b/docs/platforms/dotnet/common/tracing/instrumentation/ai-agents-module.mdx @@ -4,7 +4,7 @@ sidebar_order: 500 description: "Learn how to instrument your code to use Sentry's AI Agents module with Microsoft.Extensions.AI." --- -With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. As a prerequisite to setting up Agent Tracing with .NET, you'll need to first set up tracing. Once this is done, you can use the `Sentry.Extensions.AI` package to automatically instrument AI agents created with `Microsoft.Extensions.AI`. diff --git a/docs/platforms/javascript/common/agent-tracing-browser/index.mdx b/docs/platforms/javascript/common/agent-tracing-browser/index.mdx index 0927c1e922995..47fdec57827ae 100644 --- a/docs/platforms/javascript/common/agent-tracing-browser/index.mdx +++ b/docs/platforms/javascript/common/agent-tracing-browser/index.mdx @@ -44,7 +44,7 @@ notSupported: - javascript.capacitor --- -With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. ## Prerequisites diff --git a/docs/platforms/javascript/common/agent-tracing/index.mdx b/docs/platforms/javascript/common/agent-tracing/index.mdx index 4a9f3f7728b3c..c780427d6c285 100644 --- a/docs/platforms/javascript/common/agent-tracing/index.mdx +++ b/docs/platforms/javascript/common/agent-tracing/index.mdx @@ -30,7 +30,7 @@ supported: - javascript.nitro --- -With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. @@ -120,7 +120,7 @@ Sentry.init({ ### Streaming Gen AI Spans -`gen_ai` spans are sent as standalone envelope items instead of being bundled in the transaction by default (since SDK version `10.61.0`). This prevents AI spans with large inputs and outputs from hitting transaction payload size limits and being dropped, and is required for Conversations to work. +`gen_ai` spans are sent as standalone envelope items instead of being bundled in the transaction by default (since SDK version `10.61.0`). This prevents AI spans with large inputs and outputs from hitting transaction payload size limits and being dropped, and is required for Conversations to work. To send `gen_ai` spans as part of the transaction instead, set `streamGenAiSpans` to `false`. Self-hosted Sentry users should set this to `false`, as standalone `gen_ai` spans may not be ingested by their Sentry instance. @@ -192,7 +192,7 @@ Sentry.setConversationId(null); ## Identifying Users in Conversations -The [Conversations](/ai/observability/conversations/) view includes a **User** column. To populate it, call `setUser` once per request or session, before any AI calls: +The [Conversations](/product/agents/conversations/) view includes a **User** column. To populate it, call `setUser` once per request or session, before any AI calls: ```javascript import * as Sentry from "___SDK_PACKAGE___"; @@ -297,7 +297,7 @@ await Sentry.startSpan( -For a complete guide on naming agents across all supported frameworks, see [Naming Your Agents](/ai/observability/agents/naming/). +For a complete guide on naming agents across all supported frameworks, see [Naming Your Agents](/product/agents/naming/). diff --git a/docs/platforms/javascript/common/tracing/instrumentation/mcp-module.mdx b/docs/platforms/javascript/common/tracing/instrumentation/mcp-module.mdx index 010a32c549708..7ec8c30821680 100644 --- a/docs/platforms/javascript/common/tracing/instrumentation/mcp-module.mdx +++ b/docs/platforms/javascript/common/tracing/instrumentation/mcp-module.mdx @@ -26,7 +26,7 @@ supported: - javascript.tanstackstart-react --- -With Sentry's [MCP monitoring](/ai/observability/mcp/), you can track and debug MCP servers with full-stack context. You'll be able to monitor tool executions, prompt retrievals, resource access, and error rates. MCP monitoring data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry's [MCP monitoring](/product/agents/mcp/), you can track and debug MCP servers with full-stack context. You'll be able to monitor tool executions, prompt retrievals, resource access, and error rates. MCP monitoring data will be fully connected to your other Sentry data like logs, errors, and traces. As a prerequisite to setting up MCP monitoring with JavaScript, you'll need to first set up tracing. Once this is done, the JavaScript SDK will automatically instrument MCP servers created with supported libraries. If that doesn't fit your use case, you can use custom instrumentation described below. @@ -34,7 +34,7 @@ As a prerequisite to setting up MCP monitoring with JavaScript, you'll need to f The JavaScript SDK supports automatic instrumentation for MCP servers. We recommend adding the MCP integration to your Sentry configuration to automatically capture spans for MCP operations. -- [MCP (Model Context Protocol)](/ai/observability/mcp/getting-started/) +- [MCP (Model Context Protocol)](/product/agents/mcp/getting-started/) ## Manual Instrumentation diff --git a/docs/platforms/javascript/guides/cloudflare/features/workers-ai.mdx b/docs/platforms/javascript/guides/cloudflare/features/workers-ai.mdx index ba45e92e2149f..9d5e5c5545cb3 100644 --- a/docs/platforms/javascript/guides/cloudflare/features/workers-ai.mdx +++ b/docs/platforms/javascript/guides/cloudflare/features/workers-ai.mdx @@ -39,7 +39,7 @@ You can browse the full list of available models in the [Cloudflare Workers AI m behavior may change. -The Workers AI instrumentation does **not** infer the conversation ID automatically. To group multi-turn AI calls into a single Conversation, set it manually with `Sentry.setConversationId()`. The ID is applied as the `gen_ai.conversation.id` attribute to all AI spans within the current scope. See Tracking Conversations for details. +The Workers AI instrumentation does **not** infer the conversation ID automatically. To group multi-turn AI calls into a single Conversation, set it manually with `Sentry.setConversationId()`. The ID is applied as the `gen_ai.conversation.id` attribute to all AI spans within the current scope. See Tracking Conversations for details. When you use the [Cloudflare Agents SDK](https://developers.cloudflare.com/agents/), each agent instance already has a stable identifier, its Durable Object instance name (`this.name`), which is a natural conversation ID. @@ -114,4 +114,4 @@ export const MyChatAgent = Sentry.instrumentDurableObjectWithSentry( To unset the conversation ID, pass `null` to `Sentry.setConversationId()`. -Grouped AI calls show up in the Conversations view, where you can inspect token usage, latency, and errors across an entire conversation. +Grouped AI calls show up in the Conversations view, where you can inspect token usage, latency, and errors across an entire conversation. diff --git a/docs/platforms/php/guides/laravel/agent-tracing/index.mdx b/docs/platforms/php/guides/laravel/agent-tracing/index.mdx index c1decbaf5ec28..3f36e3380e3db 100644 --- a/docs/platforms/php/guides/laravel/agent-tracing/index.mdx +++ b/docs/platforms/php/guides/laravel/agent-tracing/index.mdx @@ -13,7 +13,7 @@ This feature is currently in Beta. Beta features are still in progress and may h -With Sentry Agent Tracing, you can monitor and debug your Laravel AI agents with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data is connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your Laravel AI agents with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data is connected to your other Sentry data like logs, errors, and traces. diff --git a/docs/platforms/python/integrations/anthropic/index.mdx b/docs/platforms/python/integrations/anthropic/index.mdx index 047d05cd6cfa5..72722592d5701 100644 --- a/docs/platforms/python/integrations/anthropic/index.mdx +++ b/docs/platforms/python/integrations/anthropic/index.mdx @@ -7,7 +7,7 @@ This integration connects Sentry with the [Anthropic Python SDK](https://github. Once you've installed this SDK, you can use the Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. -Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). ## Install @@ -54,7 +54,7 @@ def my_llm_stuff(): After running this script, the resulting data should show up in the `AI Spans` tab on the `Explore > Traces > Trace` page on Sentry.io. -If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/ai/observability/agents). +If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/integrations/google-genai/index.mdx b/docs/platforms/python/integrations/google-genai/index.mdx index 011a7592cad65..29d36017ca21c 100644 --- a/docs/platforms/python/integrations/google-genai/index.mdx +++ b/docs/platforms/python/integrations/google-genai/index.mdx @@ -7,7 +7,7 @@ This integration connects Sentry with the [Google Gen AI Python SDK](https://git Once you've installed this SDK, you can use the Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. -Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). ## Install @@ -67,7 +67,7 @@ def my_llm_stuff(): After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io. -If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/ai/observability/agents). +If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/integrations/huggingface_hub/index.mdx b/docs/platforms/python/integrations/huggingface_hub/index.mdx index 6f34bcde59bca..7ba068b35edb6 100644 --- a/docs/platforms/python/integrations/huggingface_hub/index.mdx +++ b/docs/platforms/python/integrations/huggingface_hub/index.mdx @@ -5,7 +5,7 @@ description: "Learn about using Sentry for Hugging Face Hub." This integration connects Sentry with [Hugging Face Hub](https://github.com/huggingface/huggingface_hub) in Python. -Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). ## Install @@ -87,7 +87,7 @@ if __name__ == "__main__": ``` -After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io, and in the [AI Agents Dashboard](/ai/observability/agents). +After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io, and in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/integrations/langchain/index.mdx b/docs/platforms/python/integrations/langchain/index.mdx index 13a0e2464f0a6..63bb2dcb22ea0 100644 --- a/docs/platforms/python/integrations/langchain/index.mdx +++ b/docs/platforms/python/integrations/langchain/index.mdx @@ -5,7 +5,7 @@ description: "Learn about using Sentry for LangChain." This integration connects Sentry with [LangChain](https://github.com/langchain-ai/langchain) in Python. -Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). ## Install @@ -149,7 +149,7 @@ with sentry_sdk.start_transaction(name="langchain-anthropic"): print(result) ``` -After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io, and in the [AI Agents Dashboard](/ai/observability/agents). +After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io, and in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/integrations/langgraph/index.mdx b/docs/platforms/python/integrations/langgraph/index.mdx index 3051ed71293c9..da6d7d6ea94d0 100644 --- a/docs/platforms/python/integrations/langgraph/index.mdx +++ b/docs/platforms/python/integrations/langgraph/index.mdx @@ -5,7 +5,7 @@ description: "Learn about using Sentry for LangGraph." This integration connects Sentry with [LangGraph](https://github.com/langchain-ai/langgraph) in Python. -Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). ## Install @@ -82,7 +82,7 @@ with sentry_sdk.start_transaction(name="langgraph-openai"): print(result) ``` -After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io, and in the [AI Agents Dashboard](/ai/observability/agents). +After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io, and in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/integrations/litellm/index.mdx b/docs/platforms/python/integrations/litellm/index.mdx index ee2dfec8559d2..09703cf4dbc0c 100644 --- a/docs/platforms/python/integrations/litellm/index.mdx +++ b/docs/platforms/python/integrations/litellm/index.mdx @@ -7,7 +7,7 @@ This integration connects Sentry with the [LiteLLM Python SDK](https://github.co Once you've installed this SDK, you can use the Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. -Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). ## Install @@ -71,7 +71,7 @@ print(response.choices[0].message.content) After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io. -If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/ai/observability/agents). +If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/integrations/openai/index.mdx b/docs/platforms/python/integrations/openai/index.mdx index a5a3beba374db..6579be1915cbe 100644 --- a/docs/platforms/python/integrations/openai/index.mdx +++ b/docs/platforms/python/integrations/openai/index.mdx @@ -7,7 +7,7 @@ This integration connects Sentry with the [OpenAI Python SDK](https://github.com Once you've installed this SDK, you can use Sentry Agents Tracing, a Sentry dashboard that helps you understand what's going on with your AI requests. -Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/ai/observability/agents). +Sentry AI Observability will automatically collect information about prompts, tools, tokens, and models. Learn more about the [AI Agents Dashboard](/product/agents). @@ -63,7 +63,7 @@ def my_llm_stuff(): After running this script, the resulting data should show up in the `"AI Spans"` tab on the `"Explore" > "Traces"` page on Sentry.io. -If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/ai/observability/agents). +If you manually created an Invoke Agent Span (not done in the example above) the data will also show up in the [AI Agents Dashboard](/product/agents). It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io). diff --git a/docs/platforms/python/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx b/docs/platforms/python/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx index 3f91b50d8d105..9fc7fb033dc09 100644 --- a/docs/platforms/python/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx +++ b/docs/platforms/python/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx @@ -4,7 +4,7 @@ sidebar_order: 500 description: "Learn how to manually instrument your code to use Sentry's Agents module." --- -With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. @@ -80,7 +80,7 @@ with sentry_sdk.start_span(op="gen_ai.chat", name="chat o3-mini") as span: #### Thinking / reasoning messages -Models with extended thinking (such as Anthropic's `thinking` blocks, Gemini's `thought`, or DeepSeek's `reasoning_content`) produce internal reasoning that isn't part of the user-visible reply. Represent this content as a `reasoning` part inside the assistant message, alongside the user-facing `text` part. Sentry surfaces reasoning parts separately and filters them out of the user-facing Conversations view, so don't fold thinking into a `text` part. +Models with extended thinking (such as Anthropic's `thinking` blocks, Gemini's `thought`, or DeepSeek's `reasoning_content`) produce internal reasoning that isn't part of the user-visible reply. Represent this content as a `reasoning` part inside the assistant message, alongside the user-facing `text` part. Sentry surfaces reasoning parts separately and filters them out of the user-facing Conversations view, so don't fold thinking into a `text` part. ```python import json @@ -123,7 +123,7 @@ This span represents the execution of an AI agent, capturing the full lifecycle -For a complete guide on naming agents across all supported frameworks, see [Naming Your Agents](/ai/observability/agents/naming/). +For a complete guide on naming agents across all supported frameworks, see [Naming Your Agents](/product/agents/naming/). @@ -177,7 +177,7 @@ with sentry_sdk.start_span(op="gen_ai.execute_tool", name="execute_tool get_weat Tracking Conversations has **beta** stability. Configuration options and behavior may change. -For AI applications that involve multi-turn conversations, you can use `sentry_sdk.ai.set_conversation_id()` to associate all AI spans from the same conversation. This enables you to track and analyze complete conversation flows within Sentry. +For AI applications that involve multi-turn conversations, you can use `sentry_sdk.ai.set_conversation_id()` to associate all AI spans from the same conversation. This enables you to track and analyze complete conversation flows within Sentry. The conversation ID is set as the `gen_ai.conversation.id` attribute on all AI-related spans in the current scope. To remove the conversation ID, use the `remove_conversation_id()` method on the `Scope`. @@ -208,7 +208,7 @@ response = openai.responses.create( ## Identifying Users in Conversations -The [Conversations](/ai/observability/conversations/) view includes a **User** column. To populate it, call `sentry_sdk.set_user` once per request or session, before any AI calls: +The [Conversations](/product/agents/conversations/) view includes a **User** column. To populate it, call `sentry_sdk.set_user` once per request or session, before any AI calls: ```python import sentry_sdk diff --git a/docs/platforms/python/tracing/instrumentation/custom-instrumentation/mcp-module.mdx b/docs/platforms/python/tracing/instrumentation/custom-instrumentation/mcp-module.mdx index 35ba0c533dc7b..ec9ea18567d71 100644 --- a/docs/platforms/python/tracing/instrumentation/custom-instrumentation/mcp-module.mdx +++ b/docs/platforms/python/tracing/instrumentation/custom-instrumentation/mcp-module.mdx @@ -4,7 +4,7 @@ sidebar_order: 600 description: "Learn how to manually instrument your code to use Sentry's MCP monitoring." --- -With Sentry's [MCP monitoring](/ai/observability/mcp/), you can track and debug MCP servers with full-stack context. You'll be able to monitor tool executions, prompt retrievals, resource access, and error rates. MCP monitoring data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry's [MCP monitoring](/product/agents/mcp/), you can track and debug MCP servers with full-stack context. You'll be able to monitor tool executions, prompt retrievals, resource access, and error rates. MCP monitoring data will be fully connected to your other Sentry data like logs, errors, and traces. As a prerequisite to setting up MCP monitoring with Python, you'll need to first set up tracing. Once this is done, the Python SDK will automatically instrument MCP servers created with supported libraries. If that doesn't fit your use case, you can use custom instrumentation described below. @@ -12,7 +12,7 @@ As a prerequisite to setting up MCP monitoring with Python, you'll need to first The Python SDK supports automatic instrumentation for MCP servers. We recommend adding the MCP integration to your Sentry configuration to automatically capture spans for MCP operations. -- [MCP (Model Context Protocol)](/ai/observability/mcp/getting-started/) +- [MCP (Model Context Protocol)](/product/agents/mcp/getting-started/) ## Manual Instrumentation diff --git a/docs/platforms/react-native/common/agent-tracing/index.mdx b/docs/platforms/react-native/common/agent-tracing/index.mdx index 3973e4fc063a2..41d51547b2af1 100644 --- a/docs/platforms/react-native/common/agent-tracing/index.mdx +++ b/docs/platforms/react-native/common/agent-tracing/index.mdx @@ -6,7 +6,7 @@ sidebar_section: features new: true --- -With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. ## Prerequisites diff --git a/docs/platforms/ruby/common/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx b/docs/platforms/ruby/common/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx index 33f321a43b724..d674f2b50ac60 100644 --- a/docs/platforms/ruby/common/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx +++ b/docs/platforms/ruby/common/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx @@ -4,7 +4,7 @@ sidebar_order: 500 description: "Learn how to manually instrument your code to use Sentry's Agents module." --- -With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. +With Sentry Agent Tracing, you can monitor and debug your AI systems with full-stack context. You'll be able to track key insights like token usage, latency, tool usage, and error rates. Agent Tracing data will be fully connected to your other Sentry data like logs, errors, and traces. As a prerequisite to setting up Agent Tracing with Ruby, you'll need to first set up tracing. Once this is done, you can use the custom instrumentation described below to capture AI agent spans. diff --git a/docs/ai/observability/conversations/img/conversations-detail.png b/docs/product/agents/conversations/img/conversations-detail.png similarity index 100% rename from docs/ai/observability/conversations/img/conversations-detail.png rename to docs/product/agents/conversations/img/conversations-detail.png diff --git a/docs/ai/observability/conversations/img/conversations-list.png b/docs/product/agents/conversations/img/conversations-list.png similarity index 100% rename from docs/ai/observability/conversations/img/conversations-list.png rename to docs/product/agents/conversations/img/conversations-list.png diff --git a/docs/ai/observability/conversations/index.mdx b/docs/product/agents/conversations/index.mdx similarity index 95% rename from docs/ai/observability/conversations/index.mdx rename to docs/product/agents/conversations/index.mdx index eb03564ac20e8..e26207b8f9d44 100644 --- a/docs/ai/observability/conversations/index.mdx +++ b/docs/product/agents/conversations/index.mdx @@ -1,6 +1,6 @@ --- title: Conversations -sidebar_order: 15 +sidebar_order: 30 beta: true description: >- Monitor past conversations with your AI assistants. Replay every message and @@ -28,9 +28,9 @@ You can find it in **Explore > Conversations** in the Sentry sidebar. ## Prerequisites -Conversations are built on top of [Agent Tracing](/ai/observability/agents/). Before you can use them, you need: +Conversations are built on top of [Agent Tracing](/product/agents/). Before you can use them, you need: -1. **Tracing enabled** with the Sentry SDK configured for your AI agent project. Follow the [Agent Tracing getting started guide](/ai/observability/agents/getting-started/) if you haven't already. +1. **Tracing enabled** with the Sentry SDK configured for your AI agent project. Follow the [Agent Tracing getting started guide](/product/agents/getting-started/) if you haven't already. 2. **A conversation ID on your spans.** Sentry groups spans into conversations using the `gen_ai.conversation.id` attribute. You can set this manually, or some SDK integrations infer it automatically. diff --git a/docs/ai/observability/agents/costs.mdx b/docs/product/agents/costs.mdx similarity index 95% rename from docs/ai/observability/agents/costs.mdx rename to docs/product/agents/costs.mdx index 71c6a34af706a..0f4c274668089 100644 --- a/docs/ai/observability/agents/costs.mdx +++ b/docs/product/agents/costs.mdx @@ -4,7 +4,7 @@ sidebar_order: 15 description: Learn how Sentry calculates AI model costs, where pricing data comes from, and what's not covered. --- -The **Model Cost** widget in the [AI Agents Dashboard](/ai/observability/agents/dashboards/) Models tab shows estimated costs for your LLM usage. This page explains how those costs are calculated. +The **Model Cost** widget in the [AI Agents Dashboard](/product/agents/dashboards/) Models tab shows estimated costs for your LLM usage. This page explains how those costs are calculated. ## Data Sources diff --git a/docs/ai/observability/agents/dashboards.mdx b/docs/product/agents/dashboards.mdx similarity index 79% rename from docs/ai/observability/agents/dashboards.mdx rename to docs/product/agents/dashboards.mdx index dec122a024038..0288cfe9b695d 100644 --- a/docs/ai/observability/agents/dashboards.mdx +++ b/docs/product/agents/dashboards.mdx @@ -12,7 +12,7 @@ keywords: og_image: /og-images/ai-monitoring-agents-dashboard.png --- -The AI Agents [Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=agents&sort=mostPopular) provide a comprehensive view of your AI workflows, including executions, model costs and token usage, tool calls, and recent errors. Once you've [configured the Sentry SDK](/ai/observability/agents/getting-started/) for your AI agent project, telemetry data is collected and displayed in the dashboard to support analysis of system behavior and performance. +The AI Agents [Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=agents&sort=mostPopular) provide a comprehensive view of your AI workflows, including executions, model costs and token usage, tool calls, and recent errors. Once you've [configured the Sentry SDK](/product/agents/getting-started/) for your AI agent project, telemetry data is collected and displayed in the dashboard to support analysis of system behavior and performance. Agent tracing have three dashboards: diff --git a/docs/ai/observability/agents/getting-started.mdx b/docs/product/agents/getting-started.mdx similarity index 81% rename from docs/ai/observability/agents/getting-started.mdx rename to docs/product/agents/getting-started.mdx index 3abf02cf80578..b5fdbfa8a2c85 100644 --- a/docs/ai/observability/agents/getting-started.mdx +++ b/docs/product/agents/getting-started.mdx @@ -20,7 +20,7 @@ keywords: Sentry Agent Tracing helps you track and debug AI agent applications using our supported SDKs and integrations. Monitor your complete agent workflows from user interaction to final response, including tool calls, model interactions, and custom logic. -To use [Conversations](/ai/observability/conversations/), set a conversation ID for each chat. Sentry uses the `gen_ai.conversation.id` attribute to group related AI spans, and some SDK integrations set it automatically. +To use [Conversations](/product/agents/conversations/), set a conversation ID for each chat. Sentry uses the `gen_ai.conversation.id` attribute to group related AI spans, and some SDK integrations set it automatically. To start sending AI agent data to Sentry, make sure you've created a Sentry project for your AI-enabled repository and follow one of the guides below: @@ -34,7 +34,7 @@ To start sending AI agent data to Sentry, make sure you've created a Sentry proj - [.NET](/platforms/dotnet/tracing/instrumentation/ai-agents-module/) - [Ruby](/platforms/ruby/tracing/instrumentation/custom-instrumentation/ai-agents-module/) -After setting up, [name your agents](/ai/observability/agents/naming/) so they appear as identifiable entries in the AI Agents dashboard. +After setting up, [name your agents](/product/agents/naming/) so they appear as identifiable entries in the AI Agents dashboard. diff --git a/docs/ai/observability/agents/img/advanced-scrubbing-rule.png b/docs/product/agents/img/advanced-scrubbing-rule.png similarity index 100% rename from docs/ai/observability/agents/img/advanced-scrubbing-rule.png rename to docs/product/agents/img/advanced-scrubbing-rule.png diff --git a/docs/ai/observability/agents/img/overview.png b/docs/product/agents/img/overview.png similarity index 100% rename from docs/ai/observability/agents/img/overview.png rename to docs/product/agents/img/overview.png diff --git a/docs/product/agents/index.mdx b/docs/product/agents/index.mdx new file mode 100644 index 0000000000000..8f259a7583c7c --- /dev/null +++ b/docs/product/agents/index.mdx @@ -0,0 +1,57 @@ +--- +title: Agents +sidebar_order: 215 +sidebar_section: debug +description: >- + Trace and debug AI agent workflows, conversations, and MCP servers. Monitor + agent runs, tool calls, model interactions, token usage, and errors. +keywords: + - AI agents + - AI observability + - LLM monitoring + - agent tracing + - MCP monitoring + - tool calls + - OpenAI + - Anthropic + - LangChain + - token usage + - agent tracing + - AI tracing +og_image: /og-images/ai-monitoring-agents.png +--- + +Sentry helps you understand what's going on with your AI agent workflows. It automatically collects information about agent runs, tool calls, model interactions, and errors across your entire AI pipeline—from user interaction to final response. You can also monitor MCP servers and replay past conversations. + +![Agents Overview](./img/overview.png) + +## Get Started + +To use Agents, you must have an existing Sentry account and project set up. If you don't have one, [create an account here](https://sentry.io/signup/). + +Learn how to [set up Sentry for Agents](/product/agents/getting-started/) and [name your agents](/product/agents/naming/) so they're identifiable in the dashboard. + +## Example Use Cases + +- Your AI agent is failing silently during tool execution, and you want to trace the complete agent flow to identify where it's breaking. +- Users report that your AI agent is returning unexpected or malformed responses, and you need to debug the full context of prompts, model calls, and outputs. +- Your AI agent workflows are experiencing performance issues, and you want to identify which steps (model calls, tool usage, or custom logic) are causing bottlenecks. + +## Related Features + + + + + diff --git a/docs/ai/observability/mcp/dashboard.mdx b/docs/product/agents/mcp/dashboard.mdx similarity index 86% rename from docs/ai/observability/mcp/dashboard.mdx rename to docs/product/agents/mcp/dashboard.mdx index 069fad35e8187..1137bdcd2c51e 100644 --- a/docs/ai/observability/mcp/dashboard.mdx +++ b/docs/product/agents/mcp/dashboard.mdx @@ -11,7 +11,7 @@ keywords: og_image: /og-images/ai-monitoring-mcp-dashboard.png --- -Once you've [configured the Sentry SDK](/ai/observability/mcp/getting-started/) for your MCP project, you'll start receiving data in the [MCP Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=mcp&sort=mostPopular). +Once you've [configured the Sentry SDK](/product/agents/mcp/getting-started/) for your MCP project, you'll start receiving data in the [MCP Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=mcp&sort=mostPopular). The main dashboard provides a comprehensive view of all your MCP server activities, performance metrics, and recent tool executions. diff --git a/docs/ai/observability/mcp/getting-started.mdx b/docs/product/agents/mcp/getting-started.mdx similarity index 100% rename from docs/ai/observability/mcp/getting-started.mdx rename to docs/product/agents/mcp/getting-started.mdx diff --git a/docs/ai/observability/mcp/img/overview-tab.png b/docs/product/agents/mcp/img/overview-tab.png similarity index 100% rename from docs/ai/observability/mcp/img/overview-tab.png rename to docs/product/agents/mcp/img/overview-tab.png diff --git a/docs/ai/observability/mcp/img/overview.png b/docs/product/agents/mcp/img/overview.png similarity index 100% rename from docs/ai/observability/mcp/img/overview.png rename to docs/product/agents/mcp/img/overview.png diff --git a/docs/ai/observability/mcp/index.mdx b/docs/product/agents/mcp/index.mdx similarity index 93% rename from docs/ai/observability/mcp/index.mdx rename to docs/product/agents/mcp/index.mdx index 30ec9ed2524ea..43710be2cdf62 100644 --- a/docs/ai/observability/mcp/index.mdx +++ b/docs/product/agents/mcp/index.mdx @@ -1,6 +1,6 @@ --- title: MCP Servers -sidebar_order: 20 +sidebar_order: 40 description: >- Learn how to use Sentry's MCP monitoring tools to trace and debug your Model Context Protocol implementations, including server connections, resource @@ -29,4 +29,4 @@ To use MCP Monitoring, you must have an existing Sentry account and project set ![MCP Monitoring Overview](./img/overview-tab.png) -Learn how to [set up Sentry for MCP Monitoring](/ai/observability/mcp/getting-started/). +Learn how to [set up Sentry for MCP Monitoring](/product/agents/mcp/getting-started/). diff --git a/docs/ai/observability/agents/naming.mdx b/docs/product/agents/naming.mdx similarity index 94% rename from docs/ai/observability/agents/naming.mdx rename to docs/product/agents/naming.mdx index 8b6771741f41a..d75bb1caaf551 100644 --- a/docs/ai/observability/agents/naming.mdx +++ b/docs/product/agents/naming.mdx @@ -10,7 +10,7 @@ keywords: - agent identification --- -Sentry uses the `gen_ai.agent.name` span attribute to identify agents in the [AI Agents Dashboard](/ai/observability/agents/dashboards/). Without a name, you won't be able to filter for a specific agent, group results by agent, or set up alerts for individual agents. +Sentry uses the `gen_ai.agent.name` span attribute to identify agents in the [AI Agents Dashboard](/product/agents/dashboards/). Without a name, you won't be able to filter for a specific agent, group results by agent, or set up alerts for individual agents. ## Quick Reference @@ -289,6 +289,6 @@ See [JavaScript manual instrumentation](/platforms/javascript/guides/node/agent- ## Next Steps -- [AI Agents Dashboard](/ai/observability/agents/dashboards/) — view and filter agents by name -- [Data Privacy](/ai/observability/agents/privacy/) — control what data is sent to Sentry -- [Model Costs](/ai/observability/agents/costs/) — track token usage and estimated costs +- [AI Agents Dashboard](/product/agents/dashboards/) — view and filter agents by name +- [Data Privacy](/product/agents/privacy/) — control what data is sent to Sentry +- [Model Costs](/product/agents/costs/) — track token usage and estimated costs diff --git a/docs/ai/observability/agents/privacy.mdx b/docs/product/agents/privacy.mdx similarity index 100% rename from docs/ai/observability/agents/privacy.mdx rename to docs/product/agents/privacy.mdx diff --git a/docs/ai/observability/agents/sampling.mdx b/docs/product/agents/sampling.mdx similarity index 96% rename from docs/ai/observability/agents/sampling.mdx rename to docs/product/agents/sampling.mdx index e0f9e14546caa..fca34f6ea81c8 100644 --- a/docs/ai/observability/agents/sampling.mdx +++ b/docs/product/agents/sampling.mdx @@ -191,6 +191,6 @@ sentry_sdk.logger.info( ## Next Steps -- [Getting Started](/ai/observability/agents/getting-started/) -- [Model Costs](/ai/observability/agents/costs/) -- [AI Agents Dashboard](/ai/observability/agents/dashboards/) +- [Getting Started](/product/agents/getting-started/) +- [Model Costs](/product/agents/costs/) +- [AI Agents Dashboard](/product/agents/dashboards/) diff --git a/docs/product/dashboards/sentry-dashboards/ai/agents.mdx b/docs/product/dashboards/sentry-dashboards/ai/agents.mdx index 3509d9b0ae309..d6332b0deccee 100644 --- a/docs/product/dashboards/sentry-dashboards/ai/agents.mdx +++ b/docs/product/dashboards/sentry-dashboards/ai/agents.mdx @@ -12,7 +12,7 @@ keywords: og_image: /og-images/ai-monitoring-agents-dashboard.png --- -AI Agents Dashboards, found in [Sentry Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=agents&sort=mostPopular) provide a comprehensive view of your AI workflows, including executions, model costs and token usage, tool calls, and recent errors. Once you've [configured the Sentry SDK](/ai/observability/agents/getting-started/) for your AI agent project, telemetry data is collected and displayed in the dashboard to support analysis of system behavior and performance. +AI Agents Dashboards, found in [Sentry Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=agents&sort=mostPopular) provide a comprehensive view of your AI workflows, including executions, model costs and token usage, tool calls, and recent errors. Once you've [configured the Sentry SDK](/product/agents/getting-started/) for your AI agent project, telemetry data is collected and displayed in the dashboard to support analysis of system behavior and performance. AI Agent monitoring have three dashboards: diff --git a/docs/product/dashboards/sentry-dashboards/ai/index.mdx b/docs/product/dashboards/sentry-dashboards/ai/index.mdx index 032dfc3edec99..f80a512ab6692 100644 --- a/docs/product/dashboards/sentry-dashboards/ai/index.mdx +++ b/docs/product/dashboards/sentry-dashboards/ai/index.mdx @@ -6,7 +6,7 @@ description: "Use Sentry Dashboards to monitor AI agents and MCP performance." Sentry Dashboards include dedicated views for AI monitoring. Use them to track agent executions, token usage, tool calls, and MCP (Model Context Protocol) traffic. -- **[AI Agents Dashboards](/product/dashboards/sentry-dashboards/ai/agents/)** — View AI agent workflows, model costs, token usage, and tool calls. See [Set up Agent Tracing](/ai/observability/agents/getting-started/) to get started. -- **[MCP Dashboards](/product/dashboards/sentry-dashboards/ai/mcp/)** — Monitor MCP server usage and performance. See [Set up MCP](/ai/observability/mcp/getting-started/) to get started. +- **[AI Agents Dashboards](/product/dashboards/sentry-dashboards/ai/agents/)** — View AI agent workflows, model costs, token usage, and tool calls. See [Set up Agent Tracing](/product/agents/getting-started/) to get started. +- **[MCP Dashboards](/product/dashboards/sentry-dashboards/ai/mcp/)** — Monitor MCP server usage and performance. See [Set up MCP](/product/agents/mcp/getting-started/) to get started. -For data privacy and security, see [AI Agents Data Privacy](/ai/observability/agents/privacy/). +For data privacy and security, see [AI Agents Data Privacy](/product/agents/privacy/). diff --git a/docs/product/dashboards/sentry-dashboards/ai/mcp.mdx b/docs/product/dashboards/sentry-dashboards/ai/mcp.mdx index ea511697de87b..0414223530815 100644 --- a/docs/product/dashboards/sentry-dashboards/ai/mcp.mdx +++ b/docs/product/dashboards/sentry-dashboards/ai/mcp.mdx @@ -11,7 +11,7 @@ keywords: og_image: /og-images/ai-monitoring-mcp-dashboard.png --- -MCP Dashboards, found in [Sentry Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=mcp&sort=mostPopular), provide a comprehensive view of your MCP server activities, including requests, tool executions, resource access, and prompt usage. Once you've [configured the Sentry SDK](/ai/observability/mcp/getting-started/) for your MCP project, telemetry data is collected and displayed in the dashboard to support analysis of system behavior and performance. +MCP Dashboards, found in [Sentry Dashboards](https://sentry.io/orgredirect/organizations/:orgslug/dashboards/?filter=onlyPrebuilt&query=mcp&sort=mostPopular), provide a comprehensive view of your MCP server activities, including requests, tool executions, resource access, and prompt usage. Once you've [configured the Sentry SDK](/product/agents/mcp/getting-started/) for your MCP project, telemetry data is collected and displayed in the dashboard to support analysis of system behavior and performance. MCP monitoring has four dashboards: diff --git a/docs/product/index.mdx b/docs/product/index.mdx index 8519fb52cd322..f733c8f11ca96 100644 --- a/docs/product/index.mdx +++ b/docs/product/index.mdx @@ -34,6 +34,9 @@ Follow a request from your frontend through your backend and across services. Se ### [Logs](/product/logs/) Structured logs that live alongside your errors and traces — not in a separate tool. Search by severity, filter by properties, and click from a log line straight into the trace that produced it. [Set up Logs](/product/logs/getting-started/) +### [Agents](/product/agents/) +Trace and debug AI agent workflows, conversations, and MCP servers. Sentry captures LLM calls, tool executions, token usage, and errors alongside the rest of your application telemetry. [Learn more →](/product/agents/) + ### [Session Replay](/product/session-replay/) Watch video-like reproductions of user sessions on [web](/product/session-replay/web/) and [mobile](/product/session-replay/mobile/). See exactly what a user did before, during, and after an error, including network requests, DOM events, and console messages. @@ -59,9 +62,6 @@ See the moment a release starts to degrade. Track crash-free sessions, version a ### [AI in Sentry (Seer)](/product/ai-in-sentry/) [Seer](/product/ai-in-sentry/seer/) turns your telemetry into answers. It explains why your code failed (not just where), generates merge-ready fixes with [Autofix](/product/ai-in-sentry/seer/autofix/), and catches regressions before they merge with [AI Code Review](/product/ai-in-sentry/seer/code-review/). -### [Agents Tracing](/ai/observability/agents/) -If you're building AI-powered features, Sentry traces LLM calls, agent tool executions, and token usage alongside the rest of your application telemetry. [Learn more →](/ai/observability/) - ### [Monitors and Alerts](/product/monitors-and-alerts/) Define what matters — then get notified when it breaks. Set up [cron monitors](/product/monitors-and-alerts/monitors/crons/) for scheduled jobs, [uptime monitoring](/product/monitors-and-alerts/monitors/uptime-monitoring/) for endpoints, and [alerts](/product/monitors-and-alerts/alerts/) that route to Slack, PagerDuty, or your issue tracker. @@ -70,9 +70,6 @@ Define what matters — then get notified when it breaks. Set up [cron monitors] ### [Build Distribution](/product/build-distribution/) Securely distribute app builds to your internal teams and beta testers. Upload builds from CI to streamline your distribution workflow, manage access control, and track installation analytics. -### [AI Observability](/ai/observability/) -Whether you're building AI agents or MCP servers, Sentry automatically captures performance metrics, errors, and traces across your entire AI pipeline. - ## Getting Started Sentry supports [100+ platforms and frameworks](/platforms/). Most SDKs take a few lines of code to set up: diff --git a/docs/product/sentry-basics/performance-monitoring.mdx b/docs/product/sentry-basics/performance-monitoring.mdx index 426775ead3eb4..91c5ba5f0f4ba 100644 --- a/docs/product/sentry-basics/performance-monitoring.mdx +++ b/docs/product/sentry-basics/performance-monitoring.mdx @@ -21,7 +21,7 @@ After turning on tracing, you'll automatically see [Sentry Dashboards](/product/ - [**Frontend**](/product/dashboards/sentry-dashboards/frontend/) — Core Web Vitals, transaction duration, error rates - [**Backend**](/product/dashboards/sentry-dashboards/backend/) — database queries, API calls, caches, queues, crons - [**Mobile**](/product/dashboards/sentry-dashboards/mobile/) — cold/warm app starts, TTID & TTFD, screen load times -- [**AI**](/ai/observability/agents/) — LLM call latency, token usage, tool execution performance +- [**AI**](/product/agents/) — LLM call latency, token usage, tool execution performance
diff --git a/includes/ai-dashboards/models.mdx b/includes/ai-dashboards/models.mdx index 32eb2fabfb49b..95ae10e2a8ab3 100644 --- a/includes/ai-dashboards/models.mdx +++ b/includes/ai-dashboards/models.mdx @@ -2,4 +2,4 @@ The Models dashboard displays Model Cost, Tokens Used, and Token Types widgets, ![Agents Tracing Models =900x](./img/agent-models-dash.png) -The Model Cost widget shows estimated costs based on token usage and model pricing. For details on how costs are calculated, where pricing data comes from, and what's not covered, see [Model Costs](/ai/observability/agents/costs/). +The Model Cost widget shows estimated costs based on token usage and model pricing. For details on how costs are calculated, where pricing data comes from, and what's not covered, see [Model Costs](/product/agents/costs/). diff --git a/includes/ai-dashboards/overview.mdx b/includes/ai-dashboards/overview.mdx index 08f95606dd758..1722a2182239b 100644 --- a/includes/ai-dashboards/overview.mdx +++ b/includes/ai-dashboards/overview.mdx @@ -11,7 +11,7 @@ The dashboard displays the following key widgets: - **Tokens Used**: Token usage by top models - **Tool Calls**: Tool call volume and trends -Agents are grouped by name. If your agents show up as unnamed, see [Naming Your Agents](/ai/observability/agents/naming/). +Agents are grouped by name. If your agents show up as unnamed, see [Naming Your Agents](/product/agents/naming/). Click into any model listed under widgets to get to the Models dashboard. Click into any tool listed under widgets to get to the Tools dashboard. diff --git a/includes/docs-changelog.mdx b/includes/docs-changelog.mdx index ee67a3a0ad566..45b36d4b21fb6 100644 --- a/includes/docs-changelog.mdx +++ b/includes/docs-changelog.mdx @@ -20,11 +20,11 @@ ### [docs(ai): Update AI dashboards with shared includes and new AI Spans view](https://github.com/getsentry/sentry-docs/pull/17183) **Added:** -- [AI Agents Dashboards](/ai/observability/agents/dashboards/) +- [AI Agents Dashboards](/product/agents/dashboards/) **Modified:** -- [Model Costs](/ai/observability/agents/costs/) -- [AI Agents](/ai/observability/agents/) +- [Model Costs](/product/agents/costs/) +- [AI Agents](/product/agents/) - [Instrument AI Agents](/platforms/dotnet/tracing/instrumentation/ai-agents-module/) - [Browser AI Tracing](/platforms/javascript/agent-tracing-browser/) - [Set Up Agent Tracing](/platforms/javascript/guides/node/agent-tracing/) @@ -33,7 +33,7 @@ - [AI Agents Dashboards](/product/dashboards/sentry-dashboards/ai/agents/) **Removed:** -- [/ai/observability/agents/dashboard/](/ai/observability/agents/dashboards/) +- [/product/agents/dashboards/](/product/agents/dashboards/) --- @@ -75,11 +75,11 @@ ### [docs(ai): Add naming your agents page](https://github.com/getsentry/sentry-docs/pull/17133) **Added:** -- [Naming Your Agents](/ai/observability/agents/naming/) +- [Naming Your Agents](/product/agents/naming/) **Modified:** -- [Set Up](/ai/observability/agents/getting-started/) -- [AI Agents](/ai/observability/agents/) +- [Set Up](/product/agents/getting-started/) +- [AI Agents](/product/agents/) - [Set Up Agent Tracing](/platforms/javascript/guides/node/agent-tracing/) - [Instrument AI Agents](/platforms/python/tracing/instrumentation/custom-instrumentation/ai-agents-module/) diff --git a/includes/tracing/ai-agents-module/token-cost-gotchas.mdx b/includes/tracing/ai-agents-module/token-cost-gotchas.mdx index 5cdb151bee85c..ac4f9d218a959 100644 --- a/includes/tracing/ai-agents-module/token-cost-gotchas.mdx +++ b/includes/tracing/ai-agents-module/token-cost-gotchas.mdx @@ -1,6 +1,6 @@ ## Token Usage and Cost Gotchas -When manually setting token attributes, be aware of how Sentry uses them to [calculate model costs](/ai/observability/agents/costs/). +When manually setting token attributes, be aware of how Sentry uses them to [calculate model costs](/product/agents/costs/). **Cached and reasoning tokens are subsets, not separate counts.** `gen_ai.usage.input_tokens` is the **total** input token count that already includes any cached tokens. Similarly, `gen_ai.usage.output_tokens` already includes reasoning tokens. Sentry subtracts the cached/reasoning counts from the totals to compute the "raw" portion, so reporting them incorrectly can produce wrong or negative costs. diff --git a/middleware.ts b/middleware.ts index fbb322206f84a..57b0dfa6618b9 100644 --- a/middleware.ts +++ b/middleware.ts @@ -3438,112 +3438,160 @@ const USER_DOCS_REDIRECTS: Redirect[] = [ }, { from: '/product/ai-monitoring/', - to: '/ai/observability/agents/', + to: '/product/agents/', }, { from: '/product/insights/llm-monitoring/', - to: '/ai/observability/agents/', + to: '/product/agents/', }, { from: '/product/insights/llm-monitoring/getting-started/', - to: '/ai/observability/agents/getting-started/', + to: '/product/agents/getting-started/', }, { from: '/product/insights/llm-monitoring/getting-started/the-dashboard/', - to: '/ai/observability/agents/dashboards/', + to: '/product/agents/dashboards/', }, { from: '/product/insights/ai/', - to: '/ai/observability/', + to: '/product/agents/', }, { from: '/product/insights/ai/agents/', - to: '/ai/observability/agents/', + to: '/product/agents/', }, { from: '/product/insights/ai/agents/privacy/', - to: '/ai/observability/agents/privacy/', + to: '/product/agents/privacy/', }, { from: '/product/insights/ai/agents/dashboard/', - to: '/ai/observability/agents/dashboards/', + to: '/product/agents/dashboards/', }, { from: '/ai/observability/agents/dashboard/', - to: '/ai/observability/agents/dashboards/', + to: '/product/agents/dashboards/', }, { from: '/product/insights/ai/agents/costs/', - to: '/ai/observability/agents/costs/', + to: '/product/agents/costs/', }, { from: '/product/insights/ai/agents/getting-started/', - to: '/ai/observability/agents/getting-started/', + to: '/product/agents/getting-started/', }, { from: '/product/insights/ai/mcp/', - to: '/ai/observability/mcp/', + to: '/product/agents/mcp/', }, { from: '/product/insights/ai/mcp/getting-started/', - to: '/ai/observability/mcp/getting-started/', + to: '/product/agents/mcp/getting-started/', }, { from: '/product/insights/ai/mcp/dashboard/', - to: '/ai/observability/mcp/dashboard/', + to: '/product/agents/mcp/dashboard/', }, - // AI Monitoring → AI Observability rename redirects + // AI Monitoring / AI Observability → Product Agents { from: '/ai/monitoring/', - to: '/ai/observability/', + to: '/product/agents/', }, { from: '/ai/monitoring/agents/', - to: '/ai/observability/agents/', + to: '/product/agents/', }, { from: '/ai/monitoring/agents/getting-started/', - to: '/ai/observability/agents/getting-started/', + to: '/product/agents/getting-started/', }, { from: '/ai/monitoring/agents/dashboards/', - to: '/ai/observability/agents/dashboards/', + to: '/product/agents/dashboards/', }, { from: '/ai/monitoring/agents/dashboard/', - to: '/ai/observability/agents/dashboards/', + to: '/product/agents/dashboards/', }, { from: '/ai/monitoring/agents/naming/', - to: '/ai/observability/agents/naming/', + to: '/product/agents/naming/', }, { from: '/ai/monitoring/agents/privacy/', - to: '/ai/observability/agents/privacy/', + to: '/product/agents/privacy/', }, { from: '/ai/monitoring/agents/costs/', - to: '/ai/observability/agents/costs/', + to: '/product/agents/costs/', }, { from: '/ai/monitoring/agents/sampling/', - to: '/ai/observability/agents/sampling/', + to: '/product/agents/sampling/', }, { from: '/ai/monitoring/conversations/', - to: '/ai/observability/conversations/', + to: '/product/agents/conversations/', }, { from: '/ai/monitoring/mcp/', - to: '/ai/observability/mcp/', + to: '/product/agents/mcp/', }, { from: '/ai/monitoring/mcp/getting-started/', - to: '/ai/observability/mcp/getting-started/', + to: '/product/agents/mcp/getting-started/', }, { from: '/ai/monitoring/mcp/dashboard/', - to: '/ai/observability/mcp/dashboard/', + to: '/product/agents/mcp/dashboard/', + }, + { + from: '/ai/observability/', + to: '/product/agents/', + }, + { + from: '/ai/observability/agents/', + to: '/product/agents/', + }, + { + from: '/ai/observability/agents/getting-started/', + to: '/product/agents/getting-started/', + }, + { + from: '/ai/observability/agents/dashboards/', + to: '/product/agents/dashboards/', + }, + { + from: '/ai/observability/agents/naming/', + to: '/product/agents/naming/', + }, + { + from: '/ai/observability/agents/privacy/', + to: '/product/agents/privacy/', + }, + { + from: '/ai/observability/agents/costs/', + to: '/product/agents/costs/', + }, + { + from: '/ai/observability/agents/sampling/', + to: '/product/agents/sampling/', + }, + { + from: '/ai/observability/conversations/', + to: '/product/agents/conversations/', + }, + { + from: '/ai/observability/mcp/', + to: '/product/agents/mcp/', + }, + { + from: '/ai/observability/mcp/getting-started/', + to: '/product/agents/mcp/getting-started/', + }, + { + from: '/ai/observability/mcp/dashboard/', + to: '/product/agents/mcp/dashboard/', }, { from: '/product/sentry-mcp/', diff --git a/redirects.js b/redirects.js index 5ebcf84a40cd5..22bace56eb9c5 100644 --- a/redirects.js +++ b/redirects.js @@ -1877,14 +1877,18 @@ const userDocsRedirects = [ source: '/ai/sentry-cli/', destination: 'https://cli.sentry.dev', }, - // Redirects for Insights → Sentry Dashboards migration + // Redirects for Insights → Product Agents migration + { + source: '/product/insights/llm-monitoring/', + destination: '/product/agents/', + }, { source: '/product/insights/llm-monitoring/:path*', - destination: '/ai/observability/:path*', + destination: '/product/agents/:path*', }, { source: '/product/insights/agents/:path*', - destination: '/ai/observability/agents/:path*', + destination: '/product/agents/:path*', }, { source: '/product/insights/retention-priorities/', @@ -1892,11 +1896,40 @@ const userDocsRedirects = [ }, { source: '/product/insights/mcp/', - destination: '/ai/observability/mcp/', + destination: '/product/agents/mcp/', + }, + // Flatten remaining /insights/ai/agents/* under /product/agents/* + { + source: '/product/insights/ai/agents/', + destination: '/product/agents/', + }, + { + source: '/product/insights/ai/agents/:path*', + destination: '/product/agents/:path*', + }, + { + source: '/product/insights/ai/conversations/', + destination: '/product/agents/conversations/', + }, + { + source: '/product/insights/ai/conversations/:path*', + destination: '/product/agents/conversations/:path*', + }, + { + source: '/product/insights/ai/mcp/', + destination: '/product/agents/mcp/', + }, + { + source: '/product/insights/ai/mcp/:path*', + destination: '/product/agents/mcp/:path*', + }, + { + source: '/product/insights/ai/', + destination: '/product/agents/', }, { source: '/product/insights/ai/:path*', - destination: '/ai/observability/:path*', + destination: '/product/agents/:path*', }, { source: '/product/insights/requests/', @@ -1980,14 +2013,78 @@ const userDocsRedirects = [ source: '/platforms/javascript/agent-tracing/', destination: '/platforms/javascript/agent-tracing-browser/', }, - // AI Monitoring → AI Observability rename redirects + // AI Monitoring / AI Observability → Product Agents { - source: '/ai/monitoring/:path*', - destination: '/ai/observability/:path*', + source: '/ai/monitoring/agents/dashboard/', + destination: '/product/agents/dashboards/', + }, + { + source: '/ai/monitoring/agents/', + destination: '/product/agents/', + }, + { + source: '/ai/monitoring/agents/:path*', + destination: '/product/agents/:path*', + }, + { + source: '/ai/monitoring/conversations/', + destination: '/product/agents/conversations/', + }, + { + source: '/ai/monitoring/conversations/:path*', + destination: '/product/agents/conversations/:path*', + }, + { + source: '/ai/monitoring/mcp/', + destination: '/product/agents/mcp/', + }, + { + source: '/ai/monitoring/mcp/:path*', + destination: '/product/agents/mcp/:path*', }, { source: '/ai/monitoring/', - destination: '/ai/observability/', + destination: '/product/agents/', + }, + { + source: '/ai/monitoring/:path*', + destination: '/product/agents/:path*', + }, + { + source: '/ai/observability/agents/dashboard/', + destination: '/product/agents/dashboards/', + }, + { + source: '/ai/observability/agents/', + destination: '/product/agents/', + }, + { + source: '/ai/observability/agents/:path*', + destination: '/product/agents/:path*', + }, + { + source: '/ai/observability/conversations/', + destination: '/product/agents/conversations/', + }, + { + source: '/ai/observability/conversations/:path*', + destination: '/product/agents/conversations/:path*', + }, + { + source: '/ai/observability/mcp/', + destination: '/product/agents/mcp/', + }, + { + source: '/ai/observability/mcp/:path*', + destination: '/product/agents/mcp/:path*', + }, + { + source: '/ai/observability/', + destination: '/product/agents/', + }, + { + source: '/ai/observability/:path*', + destination: '/product/agents/:path*', }, // AI Agent Monitoring / ai-agent-tracing → Agent Tracing rename redirects { diff --git a/src/components/home.tsx b/src/components/home.tsx index d6378f9ff8622..f3d0ca217d8ba 100644 --- a/src/components/home.tsx +++ b/src/components/home.tsx @@ -364,7 +364,7 @@ export async function Home() { MCP server {' '} with your agents. Debug agents and MCP servers by{' '} - + observing your AI features .