--- title: Agent Console description: >- Monitor and analyze AI agent usage, cost, and productivity across your organization in Datadog Agent Console, including coding agents, knowledge agents, and Bits AI agents. breadcrumbs: Docs > Agent Console --- > For the complete documentation index, see [llms.txt](https://docs.datadoghq.com/llms.txt). # Agent Console {% callout %} # Important note for users on the following Datadog sites: app.ddog-gov.com, us2.ddog-gov.com {% alert level="danger" %} This product is not supported for your selected [Datadog site](https://docs.datadoghq.com/getting_started/site.md). ({% placeholder "user-datadog-site-name" /%}). {% /alert %} {% /callout %} {% callout %} # Important note for users on the following Datadog sites: app.datadoghq.com, us3.datadoghq.com, us5.datadoghq.com, app.datadoghq.eu, ap1.datadoghq.com, ap2.datadoghq.com, uk1.datadoghq.com {% callout %} ##### Preview Agent Console is in Preview and available to all Datadog customers. {% /callout %} {% /callout %} The [Agent Console](https://app.datadoghq.com/llm/ai-agents-console) provides centralized monitoring for AI tools across your organization. It collects usage, cost, and activity data from coding agents, knowledge agents, desktop agents, and Datadog's own Bits AI agents. Use it to understand where your AI spend goes, who uses which tools, and how AI-assisted work affects delivery. ## Supported AI tools{% #supported-ai-tools %} | Tool | Type | Data source | | ----------------------------------------- | --------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------- | | Claude Code | Coding agent | [Anthropic Usage and Costs](https://docs.datadoghq.com/integrations/anthropic-usage-and-costs.md) integration, or OpenTelemetry for Claude Team plans | | Cursor | Coding agent | [Cursor](https://docs.datadoghq.com/integrations/cursor.md) integration | | GitHub Copilot | Coding agent | [GitHub Copilot](https://docs.datadoghq.com/integrations/github-copilot.md) integration | | Codex | Coding agent | [OpenAI Codex](https://docs.datadoghq.com/integrations/openai-codex.md) integration | | Claude Chat, Claude Cowork, Claude Design | Knowledge agent | [Anthropic Usage and Costs](https://docs.datadoghq.com/integrations/anthropic-usage-and-costs.md) integration | | ChatGPT Work | Knowledge agent | [OpenAI](https://docs.datadoghq.com/integrations/openai.md?tab=apikey) integration | | Microsoft Copilot 365 (Preview) | Knowledge agent | Microsoft Copilot integration. To request access, contact your Datadog account team. | | Other coding and desktop agents | Coding agent | [Trajectory](https://github.com/datadog-labs/trajectory) | [Trajectory](https://github.com/datadog-labs/trajectory) is an open source tool that captures sessions from more than 30 coding and desktop agents, including Claude Desktop, Cline, Windsurf, Zed, and Goose. For the full list, see [Supported Clients](https://github.com/datadog-labs/trajectory/blob/main/docs/SUPPORTED-CLIENTS.md). To connect these tools, see [Set Up Agent Console](https://docs.datadoghq.com/ai_agents_console/setup.md). ## Overview{% #overview %} The Overview tab shows your organization's AI spend, adoption, and productivity in one place. {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_overview.0a4d7535d37552cb591804f3f63304b6.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_overview.0a4d7535d37552cb591804f3f63304b6.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Agent Console Overview tab showing tracked AI cost, total users, cost per observed PR, and a breakdown of tracked cost by product area and top cost contributors" /%} The summary tiles at the top show tracked AI cost, total users, and cost per observed pull request, each compared with the previous period. The Tracked AI cost chart shows how spend is distributed across product areas, such as coding agents and other agents, and which tools contribute the most. Switch to Trend to see how cost changes over time, and select a product area to see more details. ### Productivity{% #productivity %} The Productivity section shows merged pull requests, time to merge, cost per observed pull request, and resolved Jira issues, plotted against AI cost. These metrics come from your connected [productivity tools](https://docs.datadoghq.com/ai_agents_console/setup.md#connect-productivity-tools), such as GitHub and Jira. {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_productivity_overview.317d66d648c25c933fe8151335a024ba.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_productivity_overview.317d66d648c25c933fe8151335a024ba.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Productivity section showing merged PRs, time to merge, cost per observed PR, and Jira issues resolved, each plotted against cost" /%} ### Integrations{% #integrations %} The Integrations section displays a card for each connected data source, with its status, total users, estimated cost, and token usage. Enable Group by Provider to combine products from the same provider, such as Claude Code, Chat, Design, and Cowork under Anthropic. {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_integrations_overview.01909e2a56e19a37356aa07c9e70102b.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_integrations_overview.01909e2a56e19a37356aa07c9e70102b.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Integrations section with cards for Anthropic, OpenAI, Cursor, Claude Code (OTel), GitHub Copilot, Microsoft Copilot, GitHub, and Jira" /%} Click View Dashboard on a card to open a dedicated dashboard for that integration. Filter each dashboard by team, user, and model. For example, the Anthropic dashboard breaks down estimated cost and tokens by Claude product and model: {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_integrations_anthropic.1013c198291b92989f4f8e33d8fddf90.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_integrations_anthropic.1013c198291b92989f4f8e33d8fddf90.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Anthropic dashboard showing enterprise estimated cost and tokens, cost by Claude product (Claude Code, Chat, Design, and Cowork), cost over time, and tokens by model" /%} The Codex dashboard shows estimated cost, active users, sessions, credits, and tokens, along with usage and adoption trends and top users: {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_codex_integration.7cc841f0bb1d27a640309253574dca64.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_codex_integration.7cc841f0bb1d27a640309253574dca64.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Codex dashboard showing estimated cost, active users, sessions, credits, total tokens, sessions and turns over time, tokens by type, and top users by turns and credits" /%} ## My Usage{% #my-usage %} The My Usage tab shows your own AI usage. It includes your daily spend, stacked by provider, and a breakdown of the tasks you used AI agents for. Filter by provider and model. {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_my_usage.115297c5d2b5a88333f0d4885c1f5c31.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_my_usage.115297c5d2b5a88333f0d4885c1f5c31.png?auto=format&fit=max&w=850&dpr=2 2x" alt="My Usage tab showing a personal daily spend chart stacked by provider and a task breakdown chart" /%} Use the side navigation to switch between Overview, Tasks, Productivity, Models, Skills, Tools, and Sessions. The Tasks, Skills, Tools, and Sessions views use session data from [Trajectory](https://github.com/datadog-labs/trajectory). The Models view uses data from your connected integrations. My Usage and Usage share the same views. For a description of each view, see Usage. ## Usage{% #usage %} The Usage tab provides the same views as My Usage, aggregated across your organization. Use it to understand what teams use AI agents for, identify power users and outliers, and compare adoption across teams. ### Tasks{% #tasks %} Agent Console classifies agent sessions captured with [Trajectory](https://github.com/datadog-labs/trajectory) into task categories, such as code implementation, engineering operations, and knowledge artifacts, and into more specific subcategories. The Tasks view plots each category by its share of tasks and spend per active user, and lists the cost, cost per task, and number of tasks for each one. {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_tasks.2d7c20e77b6bebbf3cd20ae120543ec7.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_tasks.2d7c20e77b6bebbf3cd20ae120543ec7.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Tasks view with a bubble chart of task categories by share of tasks and spend per active user, and a table of cost, cost per task, and task count" /%} Select a task category to open a detail panel with its subcategories, cost per task, total task cost, number of tasks, and average turns. {% image source="https://docs.dd-static.net/images/ai_agents_console/agent_console_tasks_view.c73095aa2b40722b5b4556e80ee8e8e5.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/agent_console_tasks_view.c73095aa2b40722b5b4556e80ee8e8e5.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Task category detail panel for Engineering Operations showing tasks by subcategory, cost per task, total task cost, number of tasks, and average turns" /%} ### User and team productivity{% #user-and-team-productivity %} The Productivity view gives you visibility into how individual engineers and teams use AI tooling. Use it to: - Identify your highest spenders and most productive contributors - Spot efficiency outliers, such as engineers with high spend but low output - See a full cost breakdown by user, agent, and model - Examine any individual's spend, pull request history, and model mix {% image source="https://docs.dd-static.net/images/ai_agents_console/user_analytics_user_detail_panel.6e8d57f9bdb147b95d7ee67e9f6e7428.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/user_analytics_user_detail_panel.6e8d57f9bdb147b95d7ee67e9f6e7428.png?auto=format&fit=max&w=850&dpr=2 2x" alt="User detail panel showing a breakdown for a selected user, including spend by agent, model mix, and PR history" /%} ### Models, skills, tools, and sessions{% #models-skills-tools-and-sessions %} The Models view shows which models your organization uses, based on data from your connected integrations. For agents captured with [Trajectory](https://github.com/datadog-labs/trajectory), the Skills, Tools, and Sessions views show which skills and tools agents use, and let you examine individual agent sessions. ## Bits AI agents{% #bits-ai-agents %} {% image source="https://docs.dd-static.net/images/ai_agents_console/bits_ai_agents.8875b345c141f013124812362f254d48.png?auto=format&fit=max&w=850 1x, https://docs.dd-static.net/images/ai_agents_console/bits_ai_agents.8875b345c141f013124812362f254d48.png?auto=format&fit=max&w=850&dpr=2 2x" alt="Bits AI Agents tab with a combined agent activity chart over time and individual cards for Bits Investigation, Bits Code, and Bits Agent Builder showing recent investigations, sessions, and executions" /%} The Bits AI Agents tab shows usage of Datadog's built-in AI agents alongside your other AI tools. The combined view of investigations, sessions, and executions across all Datadog agents lets you correlate Bits AI activity with the rest of your organization. Individual cards summarize activity for each Bits AI agent, including [Bits Investigation](https://docs.datadoghq.com/bits_ai/bits_investigation.md), [Bits Code](https://docs.datadoghq.com/bits_ai/bits_code.md), and [Bits Agent Builder](https://docs.datadoghq.com/actions/agents.md). Click View Details on a card to examine that agent. ## Further reading{% #further-reading %} Additional helpful documentation, links, and articles: - [Set Up Agent Console](https://docs.datadoghq.com/ai_agents_console/setup.md) - [Anthropic Usage and Costs integration](https://docs.datadoghq.com/integrations/anthropic-usage-and-costs.md) - [OpenAI Codex integration](https://docs.datadoghq.com/integrations/openai-codex.md) - [OpenAI integration](https://docs.datadoghq.com/integrations/openai.md) - [Cursor integration](https://docs.datadoghq.com/integrations/cursor.md) - [Trajectory](https://github.com/datadog-labs/trajectory) - [Monitor Claude Code adoption in your organization with Datadog's Agent Console](https://www.datadoghq.com/blog/claude-code-monitoring) - [Monitor agent adoption with Datadog Agent Console](https://www.datadoghq.com/blog/datadog-agent-console/)