TOBY AI¶

TOBY AI is the collective term for the platform's AI capabilities, including TOBY AI TruePilot for immediate personal assistance and TOBY AI Agents for continuous team work. The two use different working modes and are suited to solving different types of problems.

Before use, the workspace owner must enable the workspace AI feature under Manage > Workspace Settings > AI Analysis. This switch must be enabled first even for free features.

Product Components¶

TOBY AI TruePilot¶

TOBY AI TruePilot is provided with the workspace; no self-deployment is required; the owner must first enable the workspace AI feature. After opening it, you can describe your goal directly; on supported pages, you can also have it complete tasks based on the current object, time range, and filter conditions, using accessible data and capabilities within your existing permissions.

It is suitable for the following situations:

• Quickly understanding current logs, traces, alerts, RUM data, dashboards, or infrastructure objects;
• Writing or checking DQL, PromQL, and filtering Explorer data with natural language;
• Creating dashboard drafts or Pipeline parsing rules;
• Organizing analysis processes into Notes or Runbooks, or querying product documentation.

TOBY AI TruePilot is intended for user-initiated, on-demand tasks, helping users complete data querying, analysis, and content creation in the workspace more efficiently and accurately.

TOBY AI Agents¶

TOBY AI Agents is used to build and run Agents with clear responsibilities, enabling them to continuously participate in the team's actual work by combining workspace data, external capabilities, and the runtime environment. Each Agent can have its own responsibilities, behavioral boundaries, runtime environment, and available tools.

TOBY AI Agents is suited to complex tasks that require continuous context, multi-step collaboration, and hands-on operations, for example:

• Organizing problem analysis, solution formulation, operation execution, and result verification into continuous tasks, while maintaining context across multiple rounds of processing and cross-tool collaboration;
• Completing analysis based on observability data in the workspace and invoking code repositories, clusters, and engineering tools within the authorized scope to directly perform SRE, development, or testing work such as code fixes, scaling, and configuration adjustments;
• Connecting business systems, external services, or other AI workflows through task integrations, MCP services, and other means, bringing tasks from different sources into a unified execution pipeline;
• Configuring Agents with clear responsibilities for different teams or business scenarios to continuously execute tasks according to schedules, workspace events, or external requests, and sending results to configured Channel conversations.

An Agent's actual capabilities depend on the enabled configuration, authorization scope, and deployment environment. After creation, you can view the execution process and results in task sessions and run records.

In workspaces that support this capability, you can also directly select an Agent in the AI drawer to start a task, attach page context as needed, view progress, and open the original conversation to continue asking questions. For details, see Using Agents in the Workspace. Agents also support integrating custom models provided by the team; both model services and data access are governed by the corresponding configuration and permissions.

Quick Access¶

For instructions on opening TOBY AI TruePilot and managing conversation history, see TOBY AI TruePilot.

Plans and Credits¶

TOBY AI TruePilot and TOBY AI Agents share the workspace's AI service plan and credits account. For plan versions, credit validity, and deduction order, see Plans and Credits.